Generalized Fuzzy Torus and its Modular Properties
We consider a generalization of the basic fuzzy torus to a fuzzy torus with non-trivial modular parameter, based on a finite matrix algebra. We discuss the modular properties of this fuzzy torus, and compute the matrix Laplacian for a scalar field. In the semi-classical limit, the generalized fuzzy...
Збережено в:
| Дата: | 2013 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут математики НАН України
2013
|
| Назва видання: | Symmetry, Integrability and Geometry: Methods and Applications |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/149352 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Generalized Fuzzy Torus and its Modular Properties / P. Schreivogl, H. Steinacker // Symmetry, Integrability and Geometry: Methods and Applications. — 2013. — Т. 9. — Бібліогр.: 20 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-149352 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1493522019-02-22T01:22:54Z Generalized Fuzzy Torus and its Modular Properties Schreivogl, P. Steinacker, H. We consider a generalization of the basic fuzzy torus to a fuzzy torus with non-trivial modular parameter, based on a finite matrix algebra. We discuss the modular properties of this fuzzy torus, and compute the matrix Laplacian for a scalar field. In the semi-classical limit, the generalized fuzzy torus can be used to approximate a generic commutative torus represented by two generic vectors in the complex plane, with generic modular parameter τ. The effective classical geometry and the spectrum of the Laplacian are correctly reproduced in the limit. The spectrum of a matrix Dirac operator is also computed. 2013 Article Generalized Fuzzy Torus and its Modular Properties / P. Schreivogl, H. Steinacker // Symmetry, Integrability and Geometry: Methods and Applications. — 2013. — Т. 9. — Бібліогр.: 20 назв. — англ. 1815-0659 2010 Mathematics Subject Classification: 81R60; 81T75; 81T30 DOI: http://dx.doi.org/10.3842/SIGMA.2013.060 http://dspace.nbuv.gov.ua/handle/123456789/149352 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 |
We consider a generalization of the basic fuzzy torus to a fuzzy torus with non-trivial modular parameter, based on a finite matrix algebra. We discuss the modular properties of this fuzzy torus, and compute the matrix Laplacian for a scalar field. In the semi-classical limit, the generalized fuzzy torus can be used to approximate a generic commutative torus represented by two generic vectors in the complex plane, with generic modular parameter τ. The effective classical geometry and the spectrum of the Laplacian are correctly reproduced in the limit. The spectrum of a matrix Dirac operator is also computed. |
| format |
Article |
| author |
Schreivogl, P. Steinacker, H. |
| spellingShingle |
Schreivogl, P. Steinacker, H. Generalized Fuzzy Torus and its Modular Properties Symmetry, Integrability and Geometry: Methods and Applications |
| author_facet |
Schreivogl, P. Steinacker, H. |
| author_sort |
Schreivogl, P. |
| title |
Generalized Fuzzy Torus and its Modular Properties |
| title_short |
Generalized Fuzzy Torus and its Modular Properties |
| title_full |
Generalized Fuzzy Torus and its Modular Properties |
| title_fullStr |
Generalized Fuzzy Torus and its Modular Properties |
| title_full_unstemmed |
Generalized Fuzzy Torus and its Modular Properties |
| title_sort |
generalized fuzzy torus and its modular properties |
| publisher |
Інститут математики НАН України |
| publishDate |
2013 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/149352 |
| citation_txt |
Generalized Fuzzy Torus and its Modular Properties / P. Schreivogl, H. Steinacker // Symmetry, Integrability and Geometry: Methods and Applications. — 2013. — Т. 9. — Бібліогр.: 20 назв. — англ. |
| series |
Symmetry, Integrability and Geometry: Methods and Applications |
| work_keys_str_mv |
AT schreivoglp generalizedfuzzytorusanditsmodularproperties AT steinackerh generalizedfuzzytorusanditsmodularproperties |
| first_indexed |
2025-07-12T21:54:49Z |
| last_indexed |
2025-07-12T21:54:49Z |
| _version_ |
1837479806876254208 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 9 (2013), 060, 23 pages
Generalized Fuzzy Torus and its Modular Properties?
Paul SCHREIVOGL and Harold STEINACKER
Faculty of Physics, University of Vienna, Boltzmanngasse 5, A-1090 Vienna, Austria
E-mail: paul.schreivogl@univie.ac.at, harold.steinacker@univie.ac.at
Received June 19, 2013, in final form October 11, 2013; Published online October 17, 2013
http://dx.doi.org/10.3842/SIGMA.2013.060
Abstract. We consider a generalization of the basic fuzzy torus to a fuzzy torus with
non-trivial modular parameter, based on a finite matrix algebra. We discuss the modular
properties of this fuzzy torus, and compute the matrix Laplacian for a scalar field. In
the semi-classical limit, the generalized fuzzy torus can be used to approximate a generic
commutative torus represented by two generic vectors in the complex plane, with generic
modular parameter τ . The effective classical geometry and the spectrum of the Laplacian
are correctly reproduced in the limit. The spectrum of a matrix Dirac operator is also
computed.
Key words: fuzzy spaces; noncommutative geometry; matrix models
2010 Mathematics Subject Classification: 81R60; 81T75; 81T30
1 Introduction
In recent years, matrix models of Yang–Mills type have become a promising tool to address
fundamental questions such as the unification of interactions and gravity in physics. Their
fundamental degrees of freedom are given by a set of operators or matrices XA acting on a finite-
or infinite-dimensional Hilbert space. Specific Yang–Mills matrix models appear naturally in
string theory [5, 13], and provide a description of branes, as well as strings stretching between
the branes.
It is well-known how to realize certain basic compact branes in the framework of matrix
models. For example, the noncommutative torus T 2
θ as introduced by Connes [7] arises in certain
types matrix model compactifications, via generalized periodic boundary condition. A rich
mathematical structure has been elaborated including e.g. U-duality and Morita equivalence
of the projective modules [7, 10, 11], which is related to T-duality in string theory. However,
these results arise only due to the infinite-dimensional algebra of the non-commutative torus T 2
θ ,
which includes a non-trivial “winding sector” of string theory.
In contrast, we will focus in this paper on the class of fuzzy spaces given by the quantization
of symplectic spaces with finite symplectic volume. They arise in matrix models not via com-
pactification of but rather as embedded sub-manifolds, or “branes”. Their quantized algebra
of functions is given by a finite-dimensional simple matrix algebra AN = MN (C), without any
additional sector. As a consequence, concepts such as Morita equivalence do not make sense
a priori, and the geometry arises in a different way. A simple and well-known example is the
(rectangular) fuzzy torus T 2
N , realized in terms of finite-dimensional clock- and shift matrices.
Due to the intrinsic UV cutoff, the fuzzy tori are excellent candidates for fuzzy extra dimensions,
along the lines of [4]. The relation between T 2
N and T 2
θ was discussed in detail in [15].
As quantized symplectic manifolds, the noncommutative tori have a priori no metric struc-
ture. The infinite-dimensional noncommutative torus T 2
θ can be equipped with a differentiable
?This paper is a contribution to the Special Issue on Deformations of Space-Time and its Symmetries. The
full collection is available at http://www.emis.de/journals/SIGMA/space-time.html
mailto:paul.schreivogl@univie.ac.at
mailto:harold.steinacker@univie.ac.at
http://dx.doi.org/10.3842/SIGMA.2013.060
http://www.emis.de/journals/SIGMA/space-time.html
2 P. Schreivogl and H. Steinacker
calculus given by outer derivations, and subsequently a metric structure can be introduced via
a Laplace or Dirac operator. In contrast, the fuzzy torus T 2
N admits only inner derivations.
However if realized as brane in matrix models, it inherits an effective metric as discussed in
general in [19, 20], which is encoded in a matrix Laplace operator. This can be used to study
aspects of field theory on T 2
N [6], along the lines of the extensive literature on other fuzzy spaces
such as [1, 3, 8, 14, 16].
In this work, we study in detail the most general fuzzy torus embedded in the matrix model
as first considered in [12], and study in detail its effective geometry. We demonstrate that the
embedding provides a fuzzy analogue for a general torus with non-trivial modular parameter.
It turns out that non-trivial tori are obtained only if certain divisibility conditions for relevant
integers hold, in particularN should not be prime. In the limit of large matrices, our construction
allows to approximate any generic classical torus with generic modular parameter τ . Moreover,
we obtain a finite analogue of modular invariance, with modular group SL(2,ZN ). The effective
Riemannian and complex structure are determined using the general results in [19]. In addition
we determine the spectrum of the associated Laplace operator, and verify that the spectral
geometry is consistent with the effective geometry as determined before.
The origin for the non-trivial geometries of tori is somewhat surprising, since the embedding
in the matrix model is in a sense always rectangular. A non-rectangular effective geometry
arises due to different winding numbers along the two cycles in the apparent embedding. This
finite winding feature leads to a non-trivial modular parameter and effective metric, due to the
non-commutative nature of the branes.
This paper is organized as follows. We first review the classical results on the flat torus, as
well as the quantization of the basic rectangular fuzzy torus in the matrix model. We then give
the construction of the general fuzzy torus embedding, and determine its effective geometry.
Its modular properties are studied, and the modular group SL(2,ZN ) is identified. We also
compute the spectrum of the corresponding Laplace operator, and determine its first Brillouin
zone. Finally we also discuss the matrix Dirac operator in the rectangular case and obtain its
spectrum.
2 The classical torus
Before discussing the fuzzy torus, we review in detail the geometric structure of the classical
torus.
The most general flat 2-dimensional torus can be considered as a parallelogram in the complex
plane C, with opposite edges identified. The torus naturally inherits the metric and the complex
structure of the complex plane. The shape of the parallelogram is given by two complex num-
bers ω1 and ω2, as illustrated in Fig. 1. One can think of the vectors ω1 and ω2 as generators of
a lattice in the complex plane C. Denoting this lattice by
L(ω1, ω2) = {nω1 +mω2, n,m ∈ Z}
a point z on the torus is given by
z = σ1ω1 + σ2ω2 w σ1ω1 + σ2ω2 + 2πL(ω1, ω2),
with coordinates σ1, σ2 ∈ [0, 2π]. These points are identified according to the lattice L(ω1, ω2).
Such coordinates σ1, σ2 with periodicity 2π will be called standard coordinates. In these standard
coordinates, the line element is
ds2 =
1
2
(dzdz̄ + dz̄dz) = ω1ω̄1dσ2
1 + (ω1ω̄2 + ω2ω̄1)dσ1dσ2 + ω2ω̄2dσ2
2 = gabdσ1dσ2.
Generalized Fuzzy Torus and its Modular Properties 3
Figure 1. A torus represented as a parallelogram in the complex plane.
We can read off the metric components
gab =
(
|ω1|2 Re(ω1)Re(ω2) + Im(ω1)Im(ω2)
Re(ω1)Re(ω2) + Im(ω1)Im(ω2) |ω2|2
)
. (1)
Furthermore, we introduce the modular parameter
τ = ω1/ω2 ∈ H,
where H is the complex upper half-plane H = {z ∈ C|z > 0}. We identify conformally related
metrics on the torus. Using a Weyl scaling g → eφg of the metric as well as a diffeomorphism
(a rotation), the lattice vectors of the torus can be brought in the standard form ω1 = τ and
ω2 = 1, see Fig. 2. Then z = σ1 + τσ2 for (σ1, σ2) w (σ1, σ2) + 2π(n,m). The line element in
these standard coordinates then simplifies as
ds2 = |dσ1 + τdσ2|2,
with metric components
gab =
(
1 τ1
τ1 |τ |2
)
. (2)
In these coordinates z = σ1 + τσ2, one can express the modular parameter through the metric
components (2) as follows
τ =
g12 + i
√
g
g11
,
where g = det(gab). Now on any oriented two-dimensional Riemann surface, there is a covari-
antly constant antisymmetric tensor1 1√
g ε
ab with ε12 = −1. Together with the metric and the
1This corresponds to the inverse of the volume form.
4 P. Schreivogl and H. Steinacker
Figure 2. A torus with modular parameter τ .
antisymmetric tensor, we can build the tensor
Jab =
1
√
g
gbcε
ac. (3)
In the above standard coordinates, this tensor is explicitly
Jab =
1
τ2
2
(
τ1 −1
|τ |2 −τ1
)
and the square of J is J2 = −1. It is therefore an almost complex structure. In fact it is
a complex structure, since it is constant and thus trivially integrable.
It is instructive to choose Euclidian coordinates z = x + iy on the same torus, with metric
ds2 = dx2 + dy2. Then the periodicity becomes z w z + 2π(m+ τn). In these coordinates, the
almost complex structure takes the standard form
Jab = δbcε
ac,
which is
J =
(
0 −1
1 0
)
.
Now J2 = −1 is obvious.
Now we can discuss modular invariance. Note that two tori are always diffeomorphic as
real manifolds, but not necessarily biholomorphic as complex manifolds. This can be illustrated
e.g. with two tori T1 and T2 defined by the lattice L(ω1, ω2) = ((1, 0), (0, 1)) and L(u1, u2) =
((1, 0), (0, 2)), see Fig. 3. On T1 we choose coordinates (x1, y1), and on T2 we choose coordinates
(x2, y2). There is a diffeomorphism
(x2, y2) = (x1, 2y1).
Let us introduce complex coordinates on tori z = x1 + iy1 and w = x2 + iy2. Using the above
diffeomorphism, we obtain w = x1 + 2iy1, and together with
x1 =
z + z̄
2
, y1 =
z − z̄
2i
Generalized Fuzzy Torus and its Modular Properties 5
Figure 3. Torus T1 and T2.
we find
w =
3z − z̄
2
.
This is clearly not a holomorphic function of z.
Clearly two tori are equal as complex manifolds if their modular parameters τω = ω1/ω2 and
τu = u1/u2 coincide. Moreover, two tori are also equivalent if they are related by a modular
transformation(
a b
c d
)
∈ PSL(2,Z).
To see this, it suffices to note that the two lattices L(ω1, ω2) and L(u1, u2) are equivalent if they
are related by a PSL(2,Z) transformation(
ω1
ω2
)
=
(
a b
c d
)(
u1
u2
)
.
This leads to fractional transformation of their modular parameters
τω =
aτu + b
cτu + d
.
This modular group is in fact generated by two generators
T : τ → τ + 1, S : τ → −1/τ,
which obey the relations S2 = (ST )3 = 1. The moduli space of τ is the fundamental domain F ,
which is the complex upper half-plane H modulo the projective special linear group PSL(2,Z) =
SL(2,Z)/Z2
τ ∈ H/PSL(2,Z) = F .
A standard choice for this fundamental domain is −1/2 ≤ τ1 ≤ 1/2 and 1 ≤ |τ |, see Fig. 4.
The fundamental domain is topologically equal to the complex plane F w C. Adding the point
τ = i∞ we obtain the compactified moduli space, which is topological equivalent to the Riemann
sphere. The action of the modular transformations T : τ → τ + 1 and S : τ → −1/τ on the
torus is illustrated in Fig. 5.
6 P. Schreivogl and H. Steinacker
Figure 4. The infinite strip denoted by F is the quotient space H/PSL(2,Z) on the upper half-plane.
3 Poisson manifolds and quantization
A Poisson manifold M is a manifold together with an antisymmetric bracket {·, ·} : C(M) ×
C(M) → C(M), where C(M) denotes the space of smooth functions on M. The bracket
respects the Leibniz rule {fg, h} = f{g, h}+ g{f, h} and the Jacobi-identity {f, {g, h}}+ cyl. =
0, for f, g, h ∈ C(M). The Poisson tensor of coordinate functions is denoted as θab(x) =
{xa, xb}. If θab(x) is non-degenerate, we can introduce a symplectic form ω = 1
2θ
−1
ab dxadxb in
local coordinates. The dimension of the symplectic manifoldM is always even. The symplectic
form is closed dω = 0, which is just the Jacobi identity. Let us define a quantization map Q,
which is an isomorphism of two vector spaces. It maps the space of function to a space of
operators
Q : C(M)→ A ⊂ Mat(∞,C),
f(x)→ F.
In the present context the space of operators will be the simple matrix algebra AN = MN (C).
The quantization map Q depends on the Poisson structure, and should satisfy the conditions
Q(fg)−Q(f)Q(g)→ 0,
1
θ
(Q(i{f, g})− [Q(f),Q(g)])→ 0
for θ → 0. The algebra A is interpreted as quantized algebra of functions C(M) on M. The
quantization map Q is not unique, since higher order terms in θ are not unique. The natural
integration on symplectic manifolds
I(f) =
∫
ωn
n!
f
is related to its operator version
I(F ) = (2π)n TrF
Generalized Fuzzy Torus and its Modular Properties 7
Figure 5. The modular transformations on the torus with modular parameter τ .
in the semiclassical limit, as I(Q(f)) → I(f). Here and in the following, semiclassical limit
means taking the inverse of the quantization map Q−1(F ) = f in the limit θ → 0, keeping only
the leading contribution [·, ·] → i{·, ·} and dropping higher-order corrections in θ. Sometimes
this semi-classical limit is indicated by F → f .
We are interested here in manifolds which can be realized as Poisson manifoldM embedded
in the Euclidean space RD, with Cartesian coordinates xA, A = 1, . . . , D. The embedding is
a map
xA : M ↪→ RD,
where xA are functions on M. The Poisson tensor θab is then defined via{
xA, xB
}
= θab∂ax
A∂bx
B.
A quantization of such a Poisson manifold provides in particular quantized embedding func-
tions xA via
XA = Q
(
xA
)
∈ A ⊂ Mat(∞,C).
Now consider the action for a scalar field Φ on such a quantized Poisson manifold in the matrix
model, given by
S = −Tr
([
XA,Φ
][
XB,Φ
]
δAB
)
. (4)
In the semiclassical limit Φ ∼ φ, the action becomes
S ∼ 1
(2π)n
∫
d2nxρGab∂aφ∂bφ,
8 P. Schreivogl and H. Steinacker
where ρ =
√
det θ−1
ab . Thus Gab = θacθbdgcd is identified as effective metric. In dimensions 4
or higher, this can be cast in the standard form for a scalar field coupled to a (conformally
rescaled) metric [18]. In the present case of 2 dimensions this is not possible in general due to
Weyl invariance, cf. [2]. However we are only considering tori with constant ρ and Gab here,
where this problem is irrelevant. Then Gab = eσgab as above is indeed the effective metric, up
to possible conformal rescaling. Moreover, the matrix Laplace operator defined by
�Φ :=
[
XA,
[
XB,Φ
]]
δAB (5)
reduces in the semi-classical limit to
�Φ ∼ −gcdθacθbd∂a∂bφ = −Gab∂a∂bφ = −
√
|G| �Gφ,
where �G is the standard Laplacian on manifold with metric Gab. Thus the equation of motion
for the scalar field reduces to
�Gφ = 0
or equivalently �gφ = 0.
3.1 The rectangular fuzzy torus in the matrix model
The rectangular fuzzy torus can be defined in terms of two N × N unitary matrices, clock C
and shift S
C =
1
q
q2
. . .
qN−1
, S =
0 1 0 · · · 0
0 0 1 · · · 0
. . .
0 · · · 0 1
1 0 · · · 0
.
Here we introduce the deformation parameter q = ei2πθ, with phase θ = 1/N and positive integer
N ∈ N. The clock and shift matrices satisfy the relation
CS = qSC,
and thus
[C, S] =
(
1− q−1
)
CS.
These matrices are traceless and obey CN = SN = 1N . The fuzzy torus has a ZN × ZN
symmetry, which acts on the algebra AN as
ZN ×AN → AN ,(
ωk,Φ
)
7→ CkΦC−k
and similar for the other ZN replacing C by S. Here ω denotes the generator of ZN . Thus we
have a decomposition of the algebra of function AN over the torus into harmonics or irreducible
representations of ZN × ZN ,
AN =
N−1⊕
m,n=0
CnSm.
Generalized Fuzzy Torus and its Modular Properties 9
An element in AN can thus be written uniquely as
Φ(C, S) =
∑
|n|,|m|≤N/2
cnmq
nm
2 CnSm.
This is hermitian Φ = Φ† iff cnm = c∗−n,−m. The corresponding basis of functions on the classical
torus is einσ1eimσ2 , for n,m ∈ Z and coordinates σ1, σ2 ∈ [0, 2π]. Thus we obtain a quantization
map from the functions on the torus to a matrix algebra
Q : C
(
T 2
)
→ AN = MN (C),
einσ1eimσ2 7→
{
q
nm
2 CnSm, |n|, |m| ≤ N/2,
0, otherwise,
which is one-to-one below the UV cutoff nmax,mmax = N/2. This defines the fuzzy torus T 2
N .
Now we consider the fuzzy torus embedded in R4, via the quantized embedding functions
X1 =
R1
2
(C + C†), X2 = − iR1
2
(C − C†),
X3 =
R2
2
(S + S†), X4 = − iR2
2
(S − S†).
The hermitian matrices X1, X2, X3 and X4 satisfy the algebraic relations
X2
1 +X2
2 = R2
1, X2
3 +X2
4 = R2
2,
which tells us that R1, R2 are the radii of the torus. This embedding defines derivations given
by the adjoint action [Xi, f ] on AN .
Now consider the semi-classical limit. Then the clock and shift operators become plane waves,
C → c = eiσ1 and S → s = eiσ2 , where σa ∈ [0, 2π]. Observe that due to this periodicity, these
σa are standard coordinates on the torus as discussed before. We have then the embedding
functions xA(σ1, σ2)
x1 =
1
2
(c+ c?) = R1 cos(σ1),
x2 =
−i
2
(c− c?) = R1 sin(σ1),
x3 =
1
2
(s+ s?) = R2 cos(σ2),
x4 =
−i
2
(s− s?) = R2 sin(σ2),
which again satisfy the algebraic relations(
x1
)2
+
(
x2
)2
= R2
1,
(
x3
)2
+
(
x4
)2
= R2
2.
Using these embedding functions, we can compute the embedding (induced) metric
gab =
∂xA
∂σa
∂xB
∂σb
δAB =
(
R2
1 0
0 R2
2
)
(6)
in standard coordinates. The Poisson structure is obtained from the semiclassical limit of the
commutator
[C, S] =
(
1− q−1
)
CS → i2π
N
CS,
10 P. Schreivogl and H. Steinacker
where we expanded q to first order of 1/N . On the other hand, classically we can write for the
Poisson bracket
{c, s} = θ12∂1c∂2s = −θ12cs.
We can read off the Poisson tensor
θcd =
2π
N
(
0 −1
1 0
)
.
The corresponding symplectic structure is ω = N
π dσ1 ∧ σ2. Given the embedding metric gab
and the Poisson tensor θcd, we can compute the effective metric and the Laplacian. It is easy
to see that in 2 dimensions, the effective metric Gab = θacθbdgcd is always proportional to the
embedding metric gab by a conformal rescaling
Gab = e−σgab.
For the Laplacian in 2 dimensions such conformal factors drop out, and indeed we have always
identified conformally equivalent metrics on the torus. It is therefore sufficient here to work
only with the embedding metric gab. With these tensors at hand, we can build the complex
structure according to (3),
Jab =
θ−1
√
g
gbcθ
ca =
1
√
g
(
0 −R2
1
R2
2 0
)
,
which satisfies J2 = −1, where θ−1 = det
(
θ−1
ab
)
= N
2π . Since these are standard torus coordi-
nates, we can read off the modular parameter which is purely imaginary,
τ =
g12 + i
√
g
g11
= i
R2
R1
.
Recalling that τ = ω1/ω2, this corresponds to a rectangular torus with lattice vectors ω1 = iR2
and ω2 = R1.
3.1.1 Laplacian of a scalar field
Now consider a scalar field Φ ∈ AN on the basic fuzzy torus, with action (4)
S = −Tr
[
XA,Φ
][
XB,Φ
]
δAB
and equation of motion �Φ = 0. The matrix Laplacian operator (5) can be evaluated explicitly
on the torus as
2�Φ =
[
XA,
[
XB,Φ
]]
δAB = R2
1[C, [C†,Φ]] +R2
2[S, [S†,Φ]]
= R2
1(2Φ− CΦC† − C†ΦC) +R2
2(2Φ− SΦS† − S†ΦS),
�
(
CnSm
)
= cN
(
R2
1[n]2q +R2
2[m]2q
)
CnSm,
cN =
∣∣q1/2 − q−1/2
∣∣2 → 4π2
N2
, (7)
where we have introduced the q-number
[n]q =
qn/2 − q−n/2
q1/2 − q−1/2
=
sin(nπ/N)
sin(π/N)
→ n,
Generalized Fuzzy Torus and its Modular Properties 11
so that
[n]2q =
qn + q−n − 2
q + q−1 − 2
=
cos(2nπ/N)− 1
cos(2π/N)− 1
→ n2.
In the semiclassical limit, the spectrum2 reduces to the spectrum of the commutative Laplacian
4π2
N2
(
R2
1n
2 +R2
2m
2
)
.
4 The fuzzy torus on a general lattice
and fuzzy modular invariance
To construct more general fuzzy tori, we define two unitary operators
Vx(kx, lx) = CkxSlx , Vy(ky, ly) = CkySly , (8)
where C and S are the clock and shift matrix, and kx, lx, ky, ly ∈ Z. The operators Vx and Vy
generalize the clock and shift matrices, and satisfy V N
x = V N
y = 1. Note that the kx, lx, ky, ly
should be considered more properly as elements of ZN , due to CN = SN = 1. We combine these
kx, lx, ky, ly in two discrete complex vectors
k = kx + iky ∈ ZN + iZN ≡ CN , l = lx + ily ∈ ZN + iZN ≡ CN ,
which define a lattice
LN (k, l) = {nk +ml, n,m ∈ ZN}.
This is the fuzzy analogue of the lattice L(ω1, ω2) which defines a commutative torus. The
operators Vx(kx, lx) and Vy(ky, ly) satisfy the commutations relations
VxVy = qk∧lVyVx,
where
k ∧ l = kxly − kylx
is the area of the parallelogram spanned by k and l. Note that the operators Vx(kx, lx) and
Vy(ky, ly) commute if and only if k ∧ l = 0 mod N , corresponding to collinear vectors spanning
a degenerate torus, or tori whose area is a multiple of N .
Let us transform the lattice LN (k, l) with a PSL(2,ZN ) = SL(2,ZN )/Z2 transformation to
another lattice LN (k′, l′):(
k′
l′
)
=
(
a b
c d
)(
k
l
)
. (9)
Clearly the entries of the matrix should be elements of ZN , so that the transformed lattice
vectors k′ and l′ are in ZN . On the PSL(2,ZN ) transformed lattice LN (k′, l′) the commutation
relations are
V ′xV
′
y = qk
′∧l′V ′yV
′
x,
2It is interesting that the spectrum is the same as for a free boson in lattice theory, with lattice spacing
a = 1/N .
12 P. Schreivogl and H. Steinacker
Since the area k ∧ l is invariant under a PSL(2,ZN ) transformation
k′ ∧ l′ = (ad− bc)k ∧ l = k ∧ l,
it follows that this commutation relation is the same as for the original lattice
V ′xV
′
y = qk∧lV ′yV
′
x,
under the transformations (9). Thus we have established fuzzy modular invariance at the alge-
braic level, and we will consider noncommutative tori whose lattices are related by PSL(2,ZN )
as equal. Later we will see that the spectrum of the Laplacian and the equation of motion for
the noncommutative tori are also invariant under PSL(2,ZN ). The moduli space of the lattice
LN (k, l) or the fuzzy fundamental domain FN is defined accordingly as
FN = CN/PSL(2,ZN ). (10)
To obtain a metric structure, we define an embedding of these fuzzy tori into the R4 via the
operators Vx and Vy as follows (cf. [12])
X1 =
R1
2
(Vx + V †x ) =
R1
2
(
CkxSlx + S−lxC−kx
)
,
X2 = − iR1
2
(Vx − V †x ) = − iR1
2
(
CkxSlx − S−lxC−kx
)
,
X3 =
R2
2
(Vy + V †y ) =
R2
2
(
CkySly + S−lyC−ky
)
,
X4 = − iR2
2
(Vy − V †y ) = − iR2
2
(
CkySly − S−lyC−ky
)
. (11)
This embedding satisfies the algebraic relations X2
1 +X2
2 = R2
1 and X2
3 +X2
4 = R2
2 corresponding
to two orthogonal S1 × S1. Nevertheless, the non-trivial ansatz for the Vx,y will lead to a non-
trivial effective geometry on the tori. As usual, this embedding defines derivations on the
algebra AN given by [Xi, ·], and the integral is defined by the trace I(Φ) = 1
N Tr(Φ), where Φ
denotes a scalar field on the torus
Φ =
∑
(n1,n2)∈Z2
N
cn1n2Φn1,n2 ∈ AN , Φn1,n2 = q
n1n2
2 Cn1Sn2 .
Here the momentum space is Z2
N
∼= [−N/2 + 1, N/2]2 if N is even, to be specific. We are now
ready to compute the spectrum of the Laplacian for a scalar field on the fuzzy torus,
�LN
Φ =
[
XA,
[
XB,Φ
]]
δAB = R2
1[Vx, [V
†
x ,Φ]] +R2
2[Vy, [V
†
y ,Φ]]
= R2
1(2Φ− VxΦV †x − V †xΦVx) +R2
2(2Φ− VyΦV †y − V †y ΦVy),
�LN
(
Cn1Sn2
)
= cN (R2
1[kxn2 − lxn1]2q +R2
2[kyn2 − lyn1]2q)C
n1Sn2 =: λn1n2C
n1Sn2 .
It is easy to see that this spectrum is invariant under the SL(2,ZN ) modular transformations
acting on the defining lattice LN (k, l) as in (9), and simultaneously on the momenta as follows(
n′1
n′2
)
=
(
a b
c d
)(
n1
n2
)
.
Therefore fuzzy modular invariance is indeed a symmetry of fuzzy tori and their the scalar field
spectrum.
Generalized Fuzzy Torus and its Modular Properties 13
4.1 Spectrum and Brillouin zone
The above spectrum of �LN
has a complicated periodicity structure, and typically some dege-
neracy in momentum space CN . In order to correctly identify the irreducible spectrum and the
spectral geometry of the torus, we have to find the unit cell, or the first Brillouin zone B(~s, ~r).
This unit cell is spanned by two vectors in momentum space
~r = (r1, r2), ~s = (s1, s2) ∈ Z2
N ,
which characterize the basic periodicity of the spectrum. We can associate to them two elements
Wr = Cr1Sr2 and Ws = Cs1Ss2 in AN . Then the shift in momentum space ~n → ~n + ~r of the
field Φ along ~r is realized by ΦWr, and the shift ~n → ~n + ~s is realized by ΦWs. In order to
compute these ~s and ~r, we rewrite the spectrum in factorized form
λn1n2 = cN
(
[kxn2 − lxn1]2q + [kyn2 − lyn1]2q
)
= 4
(
1− cos
[ π
N
((kx + ky)n2 − (lx + ly)n1)
]
× cos
[ π
N
((kx − ky)n2 − (lx − ly)n1)
])
(12)
using trigonometric identities, setting R1 = R2 = 1 for simplicity. This allows to identify ~r
as primitive periodicity of the first cos factor while leaving the second unchanged, and ~s as
primitive periodicity of the second cos factor leaving the first unchanged. Explicitly,
cos
[ π
N
((kx + ky)(n2 + r2)− (lx + ly)(n1 + r1))
]
= cos
[ π
N
((kx + ky)n2 − (lx + ly)n1)
]
,
cos
[ π
N
((kx − ky)(n2 + s2)− (lx − ly)(n1 + s1))
]
= cos
[ π
N
((kx − ky)n2 − (lx − ly)n1)
]
.
This leads to the equations
(kx + ky)r2 − (lx + ly)r1 = 2N, (kx − ky)r2 − (lx − ly)r1 = 0
and
(kx + ky)s2 − (lx + ly)s1 = 0, (kx − ky)s2 − (lx − ly)s1 = 2N.
These four equations are equivalent to
kxr2 − lxr1 = N, kyr2 − lyr1 = N
and
kxs2 − lxs1 = N, kys2 − lys1 = −N,
which amount to [Vx,y,Wr,s] = 0. In complex notation, these 4 equations can be written as
kr2 − lr1 = N(1 + i), ks2 − ls1 = N(1− i)
or in matrix form(
1 + i
1− i
)
=
1
N
(
r2 −r1
s2 −s1
)(
k
l
)
. (13)
In particular, this implies
2N2 = |~r ∧ ~s||k ∧ l|, (14)
14 P. Schreivogl and H. Steinacker
reflecting the decomposition of the momentum space Z2
N into Brillouin zones. Alternatively,
these equations can be written as(
1 + i
1− i
)
=
1
N
(
kx −lx
ky −ly
)(
b
a
)
(15)
introducing the following complex combinations
a = r1 + is1, b = r2 + is2 ∈ CN .
Inverting (13) gives(
k
l
)
=
N
r1s2 − r2s1
(
−s1 r1
−s2 r2
)(
1 + i
1− i
)
. (16)
However, all quantities in these equations must be integers in [−N
2 ,
N
2 ], to be specific. Therefore
non-trivial Brillouin zones B(~s, ~r) are typically possible only if their area |~r ∧ ~s| = r1s2 − r2s1
divides3 N . Similarly, inverting (15) gives(
b
a
)
=
N
kylx − kxly
(
−ly lx
−ky kx
)(
1 + i
1− i
)
(17)
and again |k ∧ l| = kylx − kxly must typically divide N .
The above analysis leads to a very important point. The equations (17) which determine
the first Brillouin zone are Diophantic equations, so that their naive solutions in R2 may not
be admissible in CN . This follows also from (14), which is very restrictive e.g. if N is a prime
number. If (17) gives non-integer (r, s) for given (k, l), then these naive Brillouin zones and
their apparent spectral geometry are not physical; in that case, the full spectrum obtained by
properly organizing all physical modes in momentum space (n1, n2) may look very different. To
see this, consider N prime and k, l relatively prime. Then there are unitary operators C̃ = V n
x ,
S̃ = V m
y which generate AN with VxC̃ = qC̃Vx and VyS̃ = q−1C̃Vy, leading to the spectral
geometry (7) of a rectangular torus; this is in contrast to (16) which falsely suggests a non-
trivial lattice and Brillouin zone. On the other hand, if N is divisible by (kylx − kxly), then the
above equations (17) can be solved for a, b ∈ CN , for any given non-trivial lattice LN (k, l). In
that case, we obtain indeed a fuzzy version of the desired non-trivial torus as discussed below,
with periodic spectrum decomposing into several isomorphic Brillouin zones B(~s, ~r).
To illustrate this, we choose a lattice LN (k, l) with vectors l = 2 + i and k = 2 + 4i, with
area k ∧ l = 6. The smallest matrix size to accommodate this is N = 6, and in this case the
corresponding Brillouin zone B(~r,~s) is spanned by ~r = −2 + i, ~s = −6− 3i with ~r ∧ ~s = 12, see
Fig. 6. Thus momentum space decomposes into 3 copies of the Brillouin zone.
4.2 Effective geometry
Now we want to understand the effective geometry of the torus LN (k, l) in the semi-classical
limit. We will discuss both the spectral geometry as well as the effective geometry in the sense
of Section 3, which should of course agree. In the semi-classical limit, we would like that the
integers kx, lx, ky, ly approach in some sense the real numbers ω1x, ω2x, ω1y, ω2y corresponding
to some generic classical torus. More precisely, the lattice LN (k, l) should approach some given
lattice L(ω1, ω2). This can be achieved via a sequence of rational numbers approximating these
real numbers. Explicitly, we require
kN
ρN
→ ω1,
lN
ρN
→ ω2,
3This condition may be avoided e.g. if the ri, si are not relatively prime.
Generalized Fuzzy Torus and its Modular Properties 15
Figure 6. The upper parallelogram spanned by the vectors k and l is the geometric torus. The lower
parallelogram is the unit cell B(~r,~s).
where ρN is some increasing function of N . Now consider the spectrum
λn1n2 = 4 sin2
( π
N
(kxn2 − lxn1)
)
+ 4 sin2
( π
N
(kyn2 − lyn1)
)
→
(
2πρN
N
)2
|ω1n2 − ω2n1|2 (18)
setting R1 = R2 = 1. This approximation is valid as long as the argument of the sin() terms
are smaller than one, i.e. in the interior of the first Brillouin zone. As we will verify below,
this spectrum indeed reproduces the spectrum of the classical Laplace operator on the torus
L(ω1, ω2) in the semi-classical limit N →∞, as long as |ω1n2 − ω2n1| < N
ρN
.
Now consider the effective geometry in the semi-classical limit, as discussed in Section 3.
Since C ∼ eiσ1 and S ∼ eiσ2 , the defining matrices Vx and Vy (8) of the fuzzy torus LN (k, l)
become
Vx ∼ vx = ei(σ̃1ω1x+σ̃2ω2x), Vy ∼ vy = ei(σ̃1ω1y+σ̃2ω2y).
Here σ̃1,2 = ρNσi are defined on [0, 2πρN ]. The Poisson brackets can be obtained from
[Vx, Vy] ∼
2π
N
k ∧ lvxvy →
2πρ2
N
N
(ω1xω2x − ω1yω2y)vxvy.
The semi-classical approximation makes sense as long as k ∧ l < N , which holds for at least one
equivalent torus LN (k′, l′) if CN decomposes into at least N fundamental domains FN (10). We
can then identify this with the Poisson bracket
{vx, vy} = θ̃12(ω1xω2x − ω1yω2y)vxvy,
16 P. Schreivogl and H. Steinacker
and read off the Poisson tensor for the σ̃i coordinates
{
σ̃a, σ̃b
}
= θ̃ab =
2πρ2
N
N
(
0 −1
1 0
)
.
The embedding functions in R4 become
x1 =
R1
2
(vx + v?x) = R1 cos(σ̃1ω1x + σ̃2ω2x),
x2 =
−iR1
2
(vx − v?x) = R1 sin(σ̃1ω1x + σ̃2ω2x),
x3 =
R2
2
(vy + v?y) = R2 cos(σ̃1ω1y + σ̃2ω2y),
x4 =
−iR2
2
(vy − v?y) = R2 sin(σ̃1ω1y + σ̃2ω2y)
and satisfy again the algebraic relations
x2
1 + x2
2 = R2
1, x2
3 + x2
4 = R2
2.
The embedding metric is computed via (6),
ds2 =
(
(ω1xR1)2 + (ω1yR2)2
)(
dσ̃1
)2
+ 2
(
ω1xω2xR
2
1 + ω1yω2yR
2
2
)
dσ̃1dσ̃2
+
(
(ω2xR1)2 + (ω2yR2)2
)(
dσ̃2
)2
.
This reproduces indeed the metric of the general torus L(ω1, ω2) (1) for R1 = R2 = 1, which is
recovered here from a series of fuzzy tori LN (kN , lN ).
As a consistency check, we compute the spectrum of the commutative Laplacian and compare
it with the semiclassical limit (4.2). Since Gab ∼ gab in 2 dimensions as discussed before, the
Laplacian is proportional to
� = gab∂a∂b =
(
ω2
1x + ω2
1y
)
∂2
σ1 + 2(ω1xω2x + ω1yω2y)∂σ1∂σ2 +
(
ω2
2x + ω2
2y
)
∂2
σ1
setting R1 = R2 = 1 and dropping the tilde on σi. Evaluating this on einσ1eimσ2 we obtain
�einσ1eimσ2 =
[(
ω2
1x + ω2
1y
)
n2 + 2(ω1xω2x + ω1yω2y)n
2m2 +
(
ω2
2x + ω2
2y
)
m2
]
einσ1eimσ2
= |ω1m− ω2n|2einσ1eimσ2 .
This agrees (up to an irrelevant factor) with the semiclassical spectrum (4.2) of the matrix
Laplacian.
Given the metric and the Poisson structure, we can compute the complex structure
Jab =
θ̃−1
√
g
gbcθ̃
ca =
1
√
g
(
g12 −g11
g22 −g12
)
,
which satisfies J2 = −1. Here θ̃−1 = det(θ̃−1
ab ). The effective modular parameter in the commu-
tative case is given by τ = ω1/ω2 ∈ F . In the fuzzy case, we can choose a sequence of moduli
parameter depending on N
τN =
kN
lN
∈ CN ,
which for N →∞ approximates the complex number τ to arbitrary precision.
Generalized Fuzzy Torus and its Modular Properties 17
Finally let us discuss the quantization map. There is a natural map
Q : C
(
T 2
)
→ AN = MN (C),
ein1σ1ein2σ2 7→
{
q
n1n2
2 Cn1Sn2 , |ni| ≤ N
2 ,
0, otherwise,
(19)
where n1, n2 ∈ Z, and σ1, σ2 ∈ [0, 2π] are coordinates on T 2, which respects the harmonic
decomposition with respect to the classical and matrix Laplacians. In particular,
Q
(
ei(ω1xσ̃1+ω2xσ̃2)
)
= CkxSlx = Vx, Q
(
ei(ω1yσ̃1+ω2yσ̃2)
)
= CkySly = Vy
(up to phase factors) with
ω1ρN ≈ k, ω2ρN ≈ l.
Now assume that (17) is solved by integers ri, si, defining the Brillouin zone B(~r,~s). Then the
spectrum of � is n-fold degenerate, and (19) describes the quantization of an n-fold covering of
the basic torus. Indeed the elements Wr,Ws generate a discrete group GW ⊂ U(N) acting on AN
from the right, which leaves � invariant and permutes the different tori resp. Brillouin zones.
Accordingly, the space of functions on a single fuzzy torus LN (k, l) is given by the quotient
ÃN = MN (C)/GW , which is a vector space rather than an algebra. Nevertheless, it is natural
to consider the map
Q̃ : C
(
T 2
)
→ ÃN = MN (C)/GW ,
ein1σ1ein2σ2 7→
{
q
n1n2
2 Cn1Sn2 , (n1, n2) ∈ B(~r,~s),
0, otherwise,
as quantization of the torus L(ω1, ω2) under consideration.
4.3 Partition function
The partition function for a scalar field on the fuzzy torus as discussed in Section 3.1.1 is defined
via the functional approach as
ZN (k, l) =
∫
DΦe−Φ�Φ =
∫
dφnmdφn′m′e−cN
∑
nm;n′m′ φnmΩnn′;mm′φn′m′
with Qnm = [kxm−lxn]2+[kym−lyn]2 and Ωnn′;mm′ = δnn′δmm′(Qnm+ε). Here DΦ denotes the
standard measure on the space of hermitian N×N matrices, and ε is a small number introduced
to regularize the divergence due to the zero modes. The Gaussian integral gives
ZN (k, l) =
1√
det(Qnm + ε)
= ε−1/2
N−1∏
n,m 6=0
(
[kxm− lxn]2 + [kym− lyn]2 + ε
)−1/2
.
We renormalize the partition function by multiplying with ε1/2, and after taking the limit ε→ 0
we find
ZN (k, l) =
N−1∏
n,m 6=0
(
[kxm− lxn]2 + [kym− lyn]2
)−1/2
. (20)
This is completely well-defined, and invariant under the fuzzy modular group SL(2,ZN )
ZN (k′, l′) = Z(k, l)
18 P. Schreivogl and H. Steinacker
using the above results. For example, the partition function for the rectangular fuzzy torus
corresponds to the lattice ky = lx = 1 and kx = ly = 0,
ZN (1, i) =
N−1∏
n,m 6=0
(
[n]2 + [m]2
)−1/2
.
In the limit N → ∞, the partition function (20) looks very similar to the partition function of
the commutative torus L(ω1, ω2) ∼= L(τ, 1), which up to a factor takes the form
Z(ω1, ω2) =
∞∏
n,m 6=0
(
(ω1xm− ω2xn)2 + (ω1ym− ω2yn)2
)−1/2
=
∞∏
n,m 6=0
(τm+ n)(τ̄m+ n)
−1/2
.
However ZN provides a regularization which is not equivalent to a simple cutoff or zeta function
regularization (see for example [17]), because the spectrum of the fuzzy torus significantly differs
from the commutative one near the boundary of the Brillouin zone, thus regularizing the theory.
Moreover, there may be some multiplicity due to the periodic structure of Brillouin zones.
Similarly, the free energy for a scalar field on the fuzzy torus is obtained from the partition
function via
FN = lnZN = −1
2
N−1∑
n1,n2 6=0
ln
[
sin2
( π
N
(kxn2 − lxn1)
)
+ sin2
( π
N
(kyn2 − lyn1)
) ]
= −1
2
N−1∑
n1,n2 6=0
ln
[
(1− cos
( π
N
(
(kx + ky)n2 − (lx + ly)n1
))
× cos
( π
N
(
(kx − ky)n2 − (lx − ly)n1
)) ]
using the identity (12). In the semi-classical approximation
k
ρN
→ ω1,
l
ρN
→ ω2
we can replace the sum by an integral
F (ω1, ω2) = −N
2
∫
B(ω1,ω2)
dσ1dσ2 ln
[(
1− cos
(
π((ω1x + ω1y)σ1 − (ω2x + ω2y)σ2)
))
× cos
(
π((ω1x − ω1y)σ1 + (ω2x − ω2y)σ2)
)]
over the appropriate Brillouin zone, where N denotes its multiplicity. This integral is invariant
under SL(2,R) transformation of the lattice vectors ω1 and ω2. However we have not been able
to evaluate it in closed form.
We conclude with some remarks on possible applications of the above results. In the context
of string theory, a natural problem is to integrate over the moduli space of all tori. This arises
e.g. in the computation of the one-loop partition function of the bosonic string. The fuzzy torus
regularization should provide a useful new tool to address this type of problem, taking advantage
of its bounded spectrum and discretized moduli space. The integration over the moduli space
of all tori corresponds here to the sum of the partition function (20) over all fuzzy tori defined
Generalized Fuzzy Torus and its Modular Properties 19
by k and l. This is certainly finite for any given N , since the moduli space Z2
N is finite. To
define the sum over all tori, there are two natural prescriptions. First, one can consider
Z =
∑
Z2
N
ZN (k, l).
This of course entails an over-counting of lattices LN (l, k) related by SL(2,ZN ), but it is still
finite. On the other hand, one could compute
Z ′ =
∑
Z2
N/SL(2,ZN )
ZN (k, l),
which is analogous to the one-loop partition function for a closed bosonic string [17]. If all
SL(2,ZN ) orbits on Z2
N have the same cardinality, then the two definitions for Z and Z ′ are
related by a factor and hence equivalent. However this may not be true in general, and the two
definitions may not be equivalent in the large N limit. We leave a more detailed study of these
issues to future work.
Finally, the form of the spectrum of the Laplacian on LN (l, k) suggests to formulate a finite
analog of the modular form E(1, ω1, ω2)
E(1, ω1, ω2) =
∞∑
n,m 6=0
1
(ω1n+ ω2m)2
,
which could be replaced here by the fuzzy analog
Eq(1, l, k) =
∑
n,m∈B(~r,~s)\{0}
1
[kxm− lxn]2q + [kym− lyn]2q
.
This is invariant under PSL(2,ZN ), and reduces to
(
2πρN
N
)2
E(1, ω1, ω2) in the limit N →∞. It
would be interesting to construct fuzzy Eq(p, l, k) which reduce to Eisenstein series E(p, ω1, ω2)
in the limit N →∞.
4.4 The general fuzzy tori as solution of the massive matrix model
It is easy to see that the general torus corresopnding to the lattice LN as above is a solution of
the massive matrix model with equations of motion
�LN
XA = λXA (21)
as observed in [12]. Using the matrices (11) we find
�LN
XA = 4R2
i sin2
(
2π(kxly − kylx)
N
)
XA = cNR
2
i [(kxly − kylx)]2qX
A
with i = 2 for A = 1, 2 and i = 1 for A = 3, 4. Thus the embedding function Xa are solutions
of (21) for R1 = R2 = R and
cNR
2[(kxly − kylx)]2q = λ,
where cN is defined in (7). The spectrum is invariant under SL(2,ZN ) transformation, as shown
before. In the semiclassical limit, the equations of motion reduce to
�Lx
A =
(
2RρN
N
)2
(ω̄1ω2 − ω1ω̄2)2xA
or �GxA ∼ −τ2
2x
A if the lattice vectors are chosen to be ω1 = τ and ω2 = 1.
20 P. Schreivogl and H. Steinacker
4.5 Dirac operator on the fuzzy torus
In this final section we briefly discuss the Dirac equation on the rectangular fuzzy torus generated
by C and S. As usual in matrix models [5, 13, 18], the matrix Dirac operator /D is based on
the Clifford algebra of the embedding space, which is 4-dimensional here. Although this /D is in
general not equivalent to the standard Dirac operator on a Riemannian manifold, a relation can
typically be established at least in the semi-classical limit N →∞ by applying some projection
operator, as elaborated in several examples [1, 9]. Here we only study the spectrum of /D at
finite N .
First, we introduce the following representation of the two-dimensional Euclidean Gamma
matrices
γ0 =
(
0 i
−i 0
)
, γ1 =
(
0 1
1 0
)
,
which satisfy the Clifford algebra {γi, γj} = 2δij . Then a 4-dimensional Clifford algebra can
then be constructed as follows
Γ0 = γ0 ⊗
(
−1 0
0 1
)
, Γ1 = γ1 ⊗
(
−1 0
0 1
)
,
Γ2 = I ⊗
(
0 1
1 0
)
, Γ3 = I ⊗
(
0 −i
i 0
)
.
Now we define
Γ1
+ =
1
2
(
Γ0 + iΓ1
)
, Γ1
− =
1
2
(
Γ0 − iΓ1
)
,
Γ2
+ =
1
2
(
Γ2 + iΓ3
)
, Γ2
− =
1
2
(
Γ2 − iΓ3
)
.
Explicitly
Γ1
+ =
0 0 −i 0
0 0 0 i
0 0 0 0
0 0 0 0
, Γ1
− =
0 0 0 0
0 0 0 0
i 0 0 0
0 −i 0 0
,
Γ2
+ =
0 1 0 0
0 0 0 0
0 0 0 1
0 0 0 0
, Γ2
− =
0 0 0 0
1 0 0 0
0 0 0 0
0 0 1 0
.
The Dirac equation reads
/Dψ =
3∑
i=0
Γi[Xi, ψ] = λψ
or in terms of the C and S operators
/Dψ = Γ1
−[C,ψ] + Γ1
+[C†, ψ] + Γ2
−[S, ψ] + Γ2
+[S†, ψ] = λψ.
In matrix form, the Dirac operator becomes
/D =
0 [S†, ] −i[C†, ] 0
[S, ] 0 0 i[C†, ]
i[C†, ] 0 0 [S†, ]
0 −i[C, ] [S, ] 0
.
Generalized Fuzzy Torus and its Modular Properties 21
As an ansatz for a four component spinor we take
ψnm =
|n,m− 1〉anm
|n,m〉bnm
|n+ 1,m− 1〉cnm
|n+ 1,m〉dnm
,
where anm, bnm, cnm, dnm ∈ C, and |n,m〉 = CnSm ∈ AN . Using the identities
[C, |nm〉] = (1− q−m)|n+ 1,m〉, [C†, |nm〉] = (1− qm)|n− 1,m〉,
[S, |nm〉] = −(1− q−n)|n,m+ 1〉, [S†, |nm〉] = −(1− qn)|n,m− 1〉,
the Dirac equation γi[Xi, ψnm] = λnmψnm becomes explicitly
−λnm −(1− qn) −i(1− qm−1) 0
−(1− q−n) −λnm 0 i(1− qm)
i(1− q−m+1) 0 −λnm −(1− qn+1)
0 −i(1− q−m) −(1− q−n−1) −λnm
|n,m− 1〉anm
|n,m〉bnm
|n+ 1,m− 1〉cnm
|n+ 1,m〉dnm
= 0.
Setting the determinant of the matrix to zero gives
0 = λ4
nm + λ2
nm
(
− 8 + q1−m + q−1+m + q−m + qm + q−1−n + q−n + qn + q1+n
)
+
(
q−1/2−n + q1/2−m − 2q−1/2 − 2q1/2 + q−1/2+m + q1/2+n
)2
.
This can be written in terms of quadratic q-numbers
0 = λ4
nm + cNλ
2
nm
(
[1−m]2 + [m]2 + [1 + n]2 + [n]2
)
+ c2
N
(
[1/2 + n]2 − 2[1/2]2 + [1/2−m]2
)2
.
The factor cN can be absorbed by a rescaling λnm →
√
cNλnm, so that
0 = λ4
nm + λ2
nm
(
[1−m]2 + [m]2 + [1 + n]2 + [n]2
)
+
(
[1/2 + n]2 − 2[1/2]2 + [1/2−m]2
)2
.
This has four solutions, given by
λnm;1,2,3,4 = ±
{
−
(
[1−m]2 + [m]2 + [1 + n]2 + [n]2
)
±
((
[1−m]2 + [m]2 + [1 + n]2 + [n]2
)2
−
(
[1/2−m]2 + [1/2 + n]2 − 2[1/2]2
)2)1/2
}1/2
.
For the modes n,m = 0, the eigenvalues are λ00;1,2 = 0 and λ00;3,4 = ±
√
2. In the semiclassical
limit, these eigenvalues reduce to
λnm;1,2,3,4 = ±
{
−
(
− 1 +m−m2 − n− n2
)
± (1− 2m+ 2m2 + 2n+ 2n2)1/2
}1/2
.
Note that this does not and should not agree with the spectrum of the Dirac operator on
a noncommutative torus T 2
θ in the sense of [7, 15] with infinite-dimensional algebra A, since
the differential calculus here is based on inner derivations, while for T 2
θ it is based on exterior
derivations.
22 P. Schreivogl and H. Steinacker
Conclusion
We studied general fuzzy tori with algebra of functions A = MN (C) as realized in Yang–Mills
matrix models, and discussed in detail their effective geometry. Our main result is that if certain
divisibility conditions are satisfied, then the tori can have non-trivial effective geometry. The
corresponding modular space of such fuzzy tori is studied, and characterized in terms of a “fuzzy”
modular group PSL(2,ZN ). We determined the irreducible spectrum of the Laplace operator
on these tori, and exhibit their invariance under PSL(2,ZN ). In the semiclassical limit, the
general commutative torus represented by two generic vectors in the complex plane is recovered,
with generic modular parameter τ . This is quite remarkable since the “apparent” embedding is
always rectangular.
The results of this paper demonstrate the generality of the class of fuzzy embedded noncom-
mutative spaces with quantized algebra of functions A = MN (C). Moreover, our results suggest
applications of the fuzzy torus to regularize field-theoretical or string-theoretical models involv-
ing tori. A more detailed description of the moduli space (10) would be desirable, which requires
a detailed understanding of the structure of PSL(2,ZN ) for non-prime integers N . Our results
also suggest the possibility to define fuzzy analogs of modular forms. We leave an exploration
of these topics to future work.
Acknowledgments
This work was supported by the Austrian Science Fund (FWF) under the contracts P21610 and
P24713.
References
[1] Alexanian G., Balachandran A.P., Immirzi G., Ydri B., Fuzzy CP2, J. Geom. Phys. 42 (2002), 28–53,
hep-th/0103023.
[2] Arnlind J., Choe J., Hoppe J., Noncommutative minimal surfaces, arXiv:1301.0757.
[3] Arnlind J., Hoppe J., Huisken G., Discrete curvature and the Gauss–Bonnet theorem, arXiv:1001.2223.
[4] Aschieri P., Grammatikopoulos T., Steinacker H., Zoupanos G., Dynamical generation of fuzzy extra di-
mensions, dimensional reduction and symmetry breaking, J. High Energy Phys. 2006 (2006), no. 9, 026,
26 pages, hep-th/0606021.
[5] Banks T., Fischler W., Shenker S.H., Susskind L., M theory as a matrix model: a conjecture, Phys. Rev. D
55 (1997), 5112–5128, hep-th/9610043.
[6] Chaichian M., Demichev A., Prešnajder P., Sheikh-Jabbari M.M., Tureanu A., Quantum theories on non-
commutative spaces with nontrivial topology: Aharonov–Bohm and Casimir effects, Nuclear Phys. B 611
(2001), 383–402, hep-th/0101209.
[7] Connes A., Douglas M.R., Schwarz A., Noncommutative geometry and matrix theory: compactification on
tori, J. High Energy Phys. 1998 (1998), no. 2, 003, 35 pages, hep-th/9711162.
[8] Grosse H., Klimč́ık C., Prešnajder P., On finite 4D quantum field theory in non-commutative geometry,
Comm. Math. Phys. 180 (1996), 429–438, hep-th/9602115.
[9] Grosse H., Prešnajder P., The Dirac operator on the fuzzy sphere, Lett. Math. Phys. 33 (1995), 171–181.
[10] Hofman C., Verlinde E., U-duality of Born–Infeld on the noncommutative two-torus, J. High Energy Phys.
1998 (1998), no. 12, 010, 21 pages, hep-th/9810116.
[11] Hofman C., Verlinde E., Gauge bundles and Born–Infeld on the non-commutative torus, Nuclear Phys. B
547 (1999), 157–178, hep-th/9810219.
[12] Hoppe J., Some classical solutions of membrane matrix model equations, in Strings, Branes and Dualities
(Cargèse, 1997), NATO Adv. Sci. Inst. Ser. C Math. Phys. Sci., Vol. 520, Kluwer Acad. Publ., Dordrecht,
1999, 423–427, hep-th/9702169.
[13] Ishibashi N., Kawai H., Kitazawa Y., Tsuchiya A., A large-N reduced model as superstring, Nuclear Phys. B
498 (1997), 467–491, hep-th/9612115.
http://dx.doi.org/10.1016/S0393-0440(01)00070-5
http://arxiv.org/abs/hep-th/0103023
http://arxiv.org/abs/1301.0757
http://arxiv.org/abs/1001.2223
http://dx.doi.org/10.1088/1126-6708/2006/09/026
http://arxiv.org/abs/hep-th/0606021
http://dx.doi.org/10.1103/PhysRevD.55.5112
http://arxiv.org/abs/hep-th/9610043
http://dx.doi.org/10.1016/S0550-3213(01)00348-0
http://arxiv.org/abs/hep-th/0101209
http://dx.doi.org/10.1088/1126-6708/1998/02/003
http://arxiv.org/abs/hep-th/9711162
http://dx.doi.org/10.1007/BF02099720
http://arxiv.org/abs/hep-th/9602115
http://dx.doi.org/10.1007/BF00739805
http://dx.doi.org/10.1088/1126-6708/1998/12/010
http://arxiv.org/abs/hep-th/9810116
http://dx.doi.org/10.1016/S0550-3213(99)00062-0
http://arxiv.org/abs/hep-th/9810219
http://arxiv.org/abs/hep-th/9702169
http://dx.doi.org/10.1016/S0550-3213(97)00290-3
http://arxiv.org/abs/hep-th/9612115
Generalized Fuzzy Torus and its Modular Properties 23
[14] Kimura Y., Noncommutative gauge theories on fuzzy sphere and fuzzy torus from matrix model, Progr.
Theoret. Phys. 106 (2001), 445–469, hep-th/0103192.
[15] Landi G., Lizzi F., Szabo R.J., From large N matrices to the noncommutative torus, Comm. Math. Phys.
217 (2001), 181–201, hep-th/9912130.
[16] Madore J., The fuzzy sphere, Classical Quantum Gravity 9 (1992), 69–87.
[17] Nakahara M., Geometry, topology and physics, 2nd ed., Graduate Student Series in Physics, Institute of
Physics, Bristol, 2003.
[18] Steinacker H., Emergent gravity and noncommutative branes from Yang–Mills matrix models, Nuclear
Phys. B 810 (2009), 1–39, arXiv:0806.2032.
[19] Steinacker H., Emergent geometry and gravity from matrix models: an introduction, Classical Quantum
Gravity 27 (2010), 133001, 46 pages, arXiv:1003.4134.
[20] Steinacker H., Non-commutative geometry and matrix models, PoS Proc. Sci. (2011), PoS(QGQGS2011),
004, 27 pages, arXiv:1109.5521.
http://dx.doi.org/10.1143/PTP.106.445
http://dx.doi.org/10.1143/PTP.106.445
http://arxiv.org/abs/hep-th/0103192
http://dx.doi.org/10.1007/s002200000356
http://arxiv.org/abs/hep-th/9912130
http://dx.doi.org/10.1088/0264-9381/9/1/008
http://dx.doi.org/10.1201/9781420056945
http://dx.doi.org/10.1016/j.nuclphysb.2008.10.014
http://dx.doi.org/10.1016/j.nuclphysb.2008.10.014
http://arxiv.org/abs/0806.2032
http://dx.doi.org/10.1088/0264-9381/27/13/133001
http://dx.doi.org/10.1088/0264-9381/27/13/133001
http://arxiv.org/abs/1003.4134
http://arxiv.org/abs/1109.5521
1 Introduction
2 The classical torus
3 Poisson manifolds and quantization
3.1 The rectangular fuzzy torus in the matrix model
3.1.1 Laplacian of a scalar field
4 The fuzzy torus on a general lattice and fuzzy modular invariance
4.1 Spectrum and Brillouin zone
4.2 Effective geometry
4.3 Partition function
4.4 The general fuzzy tori as solution of the massive matrix model
4.5 Dirac operator on the fuzzy torus
References
|