Noncommutative Lagrange Mechanics
It is proposed how to impose a general type of ''noncommutativity'' within classical mechanics from first principles. Formulation is performed in completely alternative way, i.e. without any resort to fuzzy and/or star product philosophy, which are extensively applied within nonc...
Збережено в:
| Дата: | 2008 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут математики НАН України
2008
|
| Назва видання: | Symmetry, Integrability and Geometry: Methods and Applications |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/149039 |
| 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: | Noncommutative Lagrange Mechanics / D. Kochan // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 13 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-149039 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1490392019-02-20T01:29:05Z Noncommutative Lagrange Mechanics Kochan, D. It is proposed how to impose a general type of ''noncommutativity'' within classical mechanics from first principles. Formulation is performed in completely alternative way, i.e. without any resort to fuzzy and/or star product philosophy, which are extensively applied within noncommutative quantum theories. Newton-Lagrange noncommutative equations of motion are formulated and their properties are analyzed from the pure geometrical point of view. It is argued that the dynamical quintessence of the system consists in its kinetic energy (Riemannian metric) specifying Riemann-Levi-Civita connection and thus the inertia geodesics of the free motion. Throughout the paper, ''noncommutativity'' is considered as an internal geometric structure of the configuration space, which can not be ''observed'' per se. Manifestation of the noncommutative phenomena is mediated by the interaction of the system with noncommutative background under the consideration. The simplest model of the interaction (minimal coupling) is proposed and it is shown that guiding affine connection is modified by the quadratic analog of the Lorentz electromagnetic force (contortion term). 2008 Article Noncommutative Lagrange Mechanics / D. Kochan // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 13 назв. — англ. 1815-0659 2000 Mathematics Subject Classification: 70G45; 46L55; 53B05 http://dspace.nbuv.gov.ua/handle/123456789/149039 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 |
It is proposed how to impose a general type of ''noncommutativity'' within classical mechanics from first principles. Formulation is performed in completely alternative way, i.e. without any resort to fuzzy and/or star product philosophy, which are extensively applied within noncommutative quantum theories. Newton-Lagrange noncommutative equations of motion are formulated and their properties are analyzed from the pure geometrical point of view. It is argued that the dynamical quintessence of the system consists in its kinetic energy (Riemannian metric) specifying Riemann-Levi-Civita connection and thus the inertia geodesics of the free motion. Throughout the paper, ''noncommutativity'' is considered as an internal geometric structure of the configuration space, which can not be ''observed'' per se. Manifestation of the noncommutative phenomena is mediated by the interaction of the system with noncommutative background under the consideration. The simplest model of the interaction (minimal coupling) is proposed and it is shown that guiding affine connection is modified by the quadratic analog of the Lorentz electromagnetic force (contortion term). |
| format |
Article |
| author |
Kochan, D. |
| spellingShingle |
Kochan, D. Noncommutative Lagrange Mechanics Symmetry, Integrability and Geometry: Methods and Applications |
| author_facet |
Kochan, D. |
| author_sort |
Kochan, D. |
| title |
Noncommutative Lagrange Mechanics |
| title_short |
Noncommutative Lagrange Mechanics |
| title_full |
Noncommutative Lagrange Mechanics |
| title_fullStr |
Noncommutative Lagrange Mechanics |
| title_full_unstemmed |
Noncommutative Lagrange Mechanics |
| title_sort |
noncommutative lagrange mechanics |
| publisher |
Інститут математики НАН України |
| publishDate |
2008 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/149039 |
| citation_txt |
Noncommutative Lagrange Mechanics / D. Kochan // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 13 назв. — англ. |
| series |
Symmetry, Integrability and Geometry: Methods and Applications |
| work_keys_str_mv |
AT kochand noncommutativelagrangemechanics |
| first_indexed |
2025-07-12T20:57:15Z |
| last_indexed |
2025-07-12T20:57:15Z |
| _version_ |
1837476182283517952 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 4 (2008), 028, 9 pages
Noncommutative Lagrange Mechanics?
Denis KOCHAN †‡
† Dept. of Theoretical Physics, FMFI UK, Mlynská dolina F2, 842 48 Bratislava, Slovakia
‡ Dept. of Theoretical Physics, Nuclear Physics Institute AS CR, 250 68 Řež, Czech Republic
E-mail: denis.kochan@gmail.com
Received November 26, 2007, in final form January 29, 2008; Published online February 25, 2008
Original article is available at http://www.emis.de/journals/SIGMA/2008/028/
Abstract. It is proposed how to impose a general type of “noncommutativity” within classi-
cal mechanics from first principles. Formulation is performed in completely alternative way,
i.e. without any resort to fuzzy and/or star product philosophy, which are extensively applied
within noncommutative quantum theories. Newton–Lagrange noncommutative equations of
motion are formulated and their properties are analyzed from the pure geometrical point
of view. It is argued that the dynamical quintessence of the system consists in its kinetic
energy (Riemannian metric) specifying Riemann–Levi-Civita connection and thus the iner-
tia geodesics of the free motion. Throughout the paper, “noncommutativity” is considered
as an internal geometric structure of the configuration space, which can not be “observed”
per se. Manifestation of the noncommutative phenomena is mediated by the interaction of
the system with noncommutative background under the consideration. The simplest model
of the interaction (minimal coupling) is proposed and it is shown that guiding affine con-
nection is modified by the quadratic analog of the Lorentz electromagnetic force (contortion
term).
Key words: noncommutative mechanics; affine connection; contortion
2000 Mathematics Subject Classification: 70G45; 46L55; 53B05
Dedicated to my daughter Zoe on the occasion of her birth
1 Introduction and motivation
Physical models involving “noncommutativity” have become very popular and have been ex-
tensively studied in the last decade. Strong impact comes from string theory. It is well known
how the noncommutativity appears in D-branes, when the open string dynamics is analyzed
in the presence of constant B-field [1]. Another plumbless source of the “noncommutativity”
is pure mathematics. New mathematical ideas and structures are applicable when speculating
about the microscopical structure of the space-time. There is a common belief that this struc-
ture is smashed by some fundamental uncertainty and that precise localization of its events is
unreachable [2]. The concept of the point is lacking sense and physics is modeled on “noncom-
mutative space-time.” The mathematical language of noncommutative physics shifts from visual
geometry to rather abstract algebra. The focus of Noncommutative Geometry is nowadays really
wide. It includes quantum groups, K-theory and Fredholm Modules, (cyclic) cohomology and
index theory, deformation quantization, fuzzy geometry and so on [3].
There is a lot of articles and dissertations elaborating the ongoing noncommutative business.
Since my aim is not to trace out its history and all detailed circumstances, I refer here to survey
articles [4] that will certainly fill this gap (see also patulous reference lists therein).
Noncommutative quantum mechanics is the subject of studies at [5, 6]. The common point
of the first group of papers lies in simple replacement of the canonical Poisson brackets on the
?This paper is a contribution to the Proceedings of the 3-rd Microconference “Analytic and Algebraic Me-
thods III”. The full collection is available at http://www.emis.de/journals/SIGMA/Prague2007.html
mailto:denis.kochan@gmail.com
http://www.emis.de/journals/SIGMA/2008/028/
http://www.emis.de/journals/SIGMA/Prague2007.html
2 D. Kochan
phase space (cotangent bundle) by modified NC-brackets. Switching from Classical to Quantum
Mechanics is performed in the spirit of Dirac’s canonical quantization dictum or via the path
integral approach. However, in [6], there are proposed various field theoretical models which
yield, under some circumstances, to the effective quantum dynamics corresponding to the motion
on the noncommutative (Moyal) plane.
This article is about a similar subject, but noncommutativity is handled from the classical
perspective only. Some effort concerning this direction was done in [7]. Authors of those pa-
pers consider noncommutative classical mechanics in the fashion of symplectic and/or Poisson
geometry. They substitute the canonical two-form (Poisson brackets) on the phase space by
modified quantity, which appears in the dequantization limit ~ → 0 in the Weyl–Wigner–Moyal
star product. What they call as noncommutative mechanics is the standard classical mechanics
in the Hamilton’s picture, but with the “modified” symplectic two-form. The main objection to
such “noncommutative treatment” is that according to Darboux theorem all symplectic struc-
tures are locally equivalent, i.e. when performing a suitable change of coordinates one gets from
the canonical two-form endowing the phase space the two-form, which comes from the Weyl–
Wigner–Moyal “dequantization”. Therefore noncommutative dynamical equations derived in [7]
are ordinary Hamilton equation of motion, but they are only rewritten in different coordinates.
The only essence would be, if one would be able to canonize some class of coordinates as spe-
cial and physically privileged. But than the covariance and treasury of the coordinate free
formulation of Hamilton mechanics become broken and results start to depend on the observer.
The following paper, therefore, deals noncommutative effects in completely alternative way,
i.e. without any resorts to its Hamiltonian and/or Lagrangian precursors. Our starting point
is dynamics which is represented by classical equations of motion and its relation to an affine
connection on a underlying configuration space M . This connection can be specified by kinetic
energy which defines the system and by external forces which produce a disturbance from
the free geodesic motion. The noncommutativity is imposed as an additional internal Poisson
structure (non-constant bi-vector) on the configuration manifold M . Its presence together with
the kinetic energy gives rise to a natural nonzero contortion term. From the point of view of
“commutative RLC-connection” this term can be interpreted as a background noncommutative
“Lorentz-like” force which affects the parallel transport and thus the free motion of the system.
It will be shown that this new contortion force is always perpendicular to the actual velocity
and therefore the total mechanical energy is conserved. Moreover, one can convince him/her-
self that this extra noncommutative background force field can not be derived from a potential
energy function. Therefore the concept of Lagrangian/Hamiltonian is completely missing and
there rises a natural question what will be a reasonable quantum analog of the proposed classical
mechanics whose dynamics is governed by the Newton–Lagrange noncommutative equations of
motion. It will be clear that the Moyal–Kontsevich star product which can be considered due
to present (non-constant) Poisson structure on M does not provide satisfying answer.
2 Preliminaries
In the following section we recall some elementary geometry picturing Lagrange mechanics [8].
An autonomous mechanical system with m degrees of freedom occupies configuration space,
a m-dimensional smooth manifold M . Its dynamics is governed by the kinetic energy (non-
singular, positive) tensor g ∈ Γ(S2 T ∗M) and the co-vectorial strength field Q ∈ Γ(T ∗M) (in
general velocity dependent) describing forces that act within the system:
d
dt
(
∂g
∂ẋa
)
− ∂g
∂xa
= Qa, a = 1, . . . ,m, and g(x, ẋ) = 1
2gkl(x)ẋkẋl.
In the special case of exact strength Q = −dU (with potential U ∈ C∞(M)), one can introduce
Noncommutative Lagrange Mechanics 3
the Lagrangian function L : TM → R:(
p ∈ M,v ∈ TpM
)
∈ TM 7−→ L(p, v) := gp(v, v)− U(p) ∈ R
and write down the celebrated Euler–Lagrange equations.
Natural setting for classical mechanics is a tangent bundle of the configuration space M .
Dynamics with a given initial position and velocity is then determined by a symmetric affine
connection on M associated with the Riemannian kinetic tensor field g.
Let me remind the reader how it works. A general affine connection on M is defined as
a m-dimensional horizontal distribution on TM
τ−→ M , (p, v) τ7→ p, which is invariant under the
action of scalar multiplication in fibers: (p, v) α7→ (p, α · v) (α ∈ R). Changed into the coins:
∀ (p, v) ∈ TM T(p,v)
(
TM
)
= Ver(p,v) ⊕Hor(p,v), and Ver(p,v) ∩Hor(p,v) =
{
0
}
.
Using a local coordinate chart (xa, ẋa) on a patch of TM , one can span the space Hor(p,v) by
a collection of m vectors:
∂xa
∣∣∣∣Hor
(p,v)
:=
∂
∂xa
∣∣∣∣
(p,v)
−Ab
a(x, ẋ)
∂
∂ẋb
∣∣∣∣
(p,v)
.
Requiring the invariance w.r.t. the mentioned α-scaling: α∗
(
Hor(p,v)
)
= Hor(p,α·v), we conclude
that Ab
a(x, ẋ) = ẋcΓ b
ca(x). Here Γ stands for a set of m3 Christoffel symbols.
Prescribing an affine connection on M , one is able to perform a horizontal lift of smooth
curve γ : [0, 1] → M to a tangent bundle curve γ̂ : [0, 1] → TM . Concisely, suppose that at
a given time t we are occupying the tangent bundle point γ̂(t) = (p, v) and a tangent vector
to γ(t) at p = τ(p, v) is γ̇(p) ∈ TpM . Then, after lapsing an infinitesimally short time ε, the new
tangent bundle position γ̂(t + ε) will be specified by an ε-step in the direction of the horizontal
lift γ̇(p)
∣∣Hor
(p,v)
. In other words, in coordinate patch (xa, ẋa) we have the following system of
coupled ordinary differential equations for γ̂:
xa(t + ε) = xa(t) + εγ̇a(t), ẋa(t + ε) = ẋa(t)− εγ̇c(t) ẋb(t)Γ a
bc
(
x(t)
)
. (1)
Here γ̇a(t) stands for the components of the tangent vector γ̇(t) ∈ Tγ(t)M to γ with respect to
the base coordinate basis {∂xa |γ(t)}.
Specifying the starting point for γ̂ in TM , i.e. the initial conditions for the above differential
system:{
xa(0) = xa(γ(0)), ẋa(0) = va
}
⇐⇒ v = va∂xa
∣∣∣
γ(0)
we are able to perform a parallel transport of the vector v ∈ Tγ(0)M along γ. Parallel transported
vector is the solution of (1) at the final time t = 1, i.e. ẋa(1)∂xa |γ(1).
For a fixed connection, there is a special class of curves that are called geodesics. Curve
γ : [0, 1] → M is a geodesic line on M , if parallel transport of γ̇(0) along γ coincides at each
time t with its actual tangent vector γ̇(t). In other words, being in a local chart on M , the
geodesic γ satisfies the system of second order differential equations:
ẍa
(
γ(t)
)
= −ẋc
(
γ(t)
)
ẋb
(
γ(t)
)
Γ a
bc
(
x
(
γ(t)
))
.
With any affine connection one can associate its torsion and curvature (for more details
see [9]). Torsion of two vectors u, v ∈ TpM is a vector Tp(u, v) at point p, which closes an
infinitesimal geodesic parallelogram framed on u, v up to second order. Thus
Tp(u, v) = ubvcTa
bc(p)∂xa
∣∣∣
p
= ubvc
{
Γ a
cb − Γ a
bc
}
(p)∂xa
∣∣∣
p
.
4 D. Kochan
It is clear that torsion is an element of Γ
(∧2 T ∗M ⊗ TM
)
, i.e. it is skew-symmetric in the
subscript indices.
Curvature associated with tangent vectors u, v ∈ TpM is a linear map, (1, 1)-tensor, Rp(u, v) :
TpM → TpM . Again, analogically as before, the curvature measures how a vector w changes,
up to second order, after parallel transport along an infinitesimal closed parallelogram spanned
on vectors u and v. Concisely,
Rp (u, v)(w) = wbucvd Ra
bcd(p)∂xa
∣∣∣
p
= wbucvd
{
Γ a
bd,c − Γ a
bc,d + Γ i
bdΓ
a
ic − Γ i
bcΓ
a
id
}
(p)∂xa
∣∣∣
p
.
We shall now describe how this is related to the Lagrange mechanics. The kinetic energy
tensor g, due to its symmetry, can be equivalently interpreted as a quadratic function on TM
w.r.t. fiber (doted) coordinates. Using a local chart, one has:
g = 1
2gab(x)ẋaẋb.
Lets define the horizontal distribution at tangent bundle point (p , v) by
Hor(p,v) :=
{
w
∣∣
(p,v)
= wa∂xa
∣∣Hor
(p,v)
∈ T(p,v)(TM); w
∣∣
(p,v)
(g) = 0
}
.
Requiring also that it is torsionless, one gets a system of equations for the Riemann–Levi-Civita
connection:
∂xc(gab) ≡ gab,c = gaiΓ
i
bc + gbiΓ
i
ac, 0 = Γ a
cb − Γ a
bc.
Regularity of the (m×m) matrix g ensures the unique solution of the above algebraic equations:
Γ a
bc = 1
2 gai
{
gib, c + gic, b − gbc, i
}
≡
{ a
b c
}
RLC
and the existence of the canonical (musical1) isomorphism ]g : T ∗M → TM . Therefore the
external strength described by the co-vectorial field Q = Qa(x, ẋ)dxa can be turned into a force
vector field ]g(Q) = (gabQb)∂a = F a∂a = F .
Suppose that a mechanical system at time t is surrounded at given point p ∈ M and possesses
the velocity v ∈ TpM . Then the new position and velocity after the infinitesimal time interval ε
is determined by the Lagrange dynamical vector field evaluated at (p, v)
L
∣∣∣
(p,v)
= v
∣∣∣Hor
(p,v)
+ F
∣∣∣Ver
(p,v)
= va∂xa
∣∣∣Hor
(p,v)
+ F a(p , v)∂ẋa
∣∣∣
(p,v)
∈ T(p,v)(TM). (2)
Expressing this dynamics in local coordinates (xa, ẋa), one gets the following system of diffe-
rential equations:
xa(t + ε) = xa(t) + εẋa(t),
ẋa(t + ε) = ẋa(t)− εẋc(t)ẋb(t)Γ a
bc
(
x(t)
)
+ εF a
(
x(t), ẋ(t)
)
.
It is straightforward to see that these equations are exactly the Euler–Lagrange equations from
the beginning of this introductory paragraph. One just needs to replace the general Γ a
bc by the
RLC Christoffel symbols
{
a
b c
}
RLC
.
1]g symbolizes the index raising with the help of dual (inverse) matrix gai(·gib = δa
b ), for the index lowering
with gai there is symbol [g.
Noncommutative Lagrange Mechanics 5
3 Noncommutative Lagrange mechanics
In the previous section we have seen that the mechanical system would evolve geodesically. The
only disturbance of such motion is caused by the presence of external forces. Thus we can
conclude that the pivotal object defining the “theory” consists in its internal kinetic energy
tensor g (lets call it a metric). States of the mechanical system are labeled by points of the
tangent bundle of the underlying configuration space M (no internal degrees of freedom are
considered).
“Noncommutativity” is an internal property of M coming a priori from the Nature. In
general it is specified by nonconstant Poisson brackets defined on C∞(M), i.e. by a certain
Poisson bi-vector field Π ∈ Γ(
∧2 TM). Let us recall that locally:
Π = 1
2Πab(x)
{
∂xa ⊗ ∂xb − ∂xb ⊗ ∂xa
}
⇐⇒ {f,h} := Π(df, dh) ≡ f,aΠabh,b.
In terms of the bi-vector Π the Jacobi identity is equivalent to the vanishing of the Schouten
brackets of the bi-vector Π with itself, i.e.
0 = [Π,Π]
Schouten
⇐⇒
{
Πai(Πbc),i + cyclic permutation in (a, b, c)
}
= 0.
The presence of the Poisson structure on M enables us to quantize the algebra of func-
tions on M (see [10]). Quantization is a formal one parameter deformation of the ordinary
pointwise product on C∞(M). Resulting noncommutative associative star-product algebra(
C∞(M)[[~]], ?~
)
satisfies (in the semiclassical regime):
lim
~→0
1
~
(
f ?~ h− h ?~ f
)
= Π(df, dh).
What physicists are quite often doing when discussing the noncommutative theories (see e.g. [4]),
is a replacement of the ordinary pointwise product in the governing action (Lagrangian density)
by an appropriate star-product ?~ and going to quantum theory. The techniques and dictionary
used for analyzing of the appearing NC-problems are classical (commutative) ones.
If the noncommutativity is manifestly present in Nature, then it should somehow more con-
ceptually affect matters also on the classical level. Not only in partial changes in the guiding
Lagrangian and/or Hamiltonian, but directly, modifying the classical equations of motion. These
are rather more fundamental than the Lagrangian and/or Hamiltonian themselves. Therefore
what I am going to do below, is to pursue this direction.
Suppose that the classical “theory” is specified by the metric g and occupies the noncom-
mutative configuration space (M,Π). Can one use those ingredients to create something new?
Let us consider the (1, 1)-tensor lowering one index of Π with the help of g (this I called the
minimal coupling of the “theory” with the “noncommutativity”), i.e.
Contr(g ⊗Π) = dxagai(x)Πib(x)∂xb = dxaΠb
a(x)∂xb = [gΠ.
With any (1, 1)-tensor A one can associate the special (1, 2)-tensor called its Nijenhuis tor-
sion [11]. It is defined as follows:
NA(X, Y ) :=
[
A(X), A(Y )
]
−A
([
A(X), Y
]
+
[
X, A(Y )
]
−A
([
X, Y
]))
.
Here the arguments X and Y are arbitrary vector fields on M and [ , ] stands for their commu-
tator. Checking C∞(M)-linearity and exhibiting its properties w.r.t. argument interchange, one
immediately concludes that NA ∈ Γ(
∧2 T ∗M ⊗ TM), i.e. it is skew-symmetric in the subscript
co-vectorial indices.
6 D. Kochan
What do we get when substituting [gΠ for A? Nijenhuis torsion of [gΠ is expressed in the
local coordinate chart on M as follows:
N[gΠ = dxb ∧ dxc 1
2
{
Πi
b(Π
a
c ),i + Πa
i (Π
i
b),c
}
∂xa . (3)
The (1, 2)-tensor N[gΠ appears on M very naturally. It reflects the properties of both entering
entities: “theory” g and “noncommutativity” Π. Therefore it is quite reasonable to postulate:
the affine connection for the NC-classical mechanics is the metric g connection, whose torsion
is N[gΠ. This asks us to solve the modified Riemann–Levi-Civita equations:
gab, c = gaiΓ
i
bc + gbiΓ
i
ac, Ta
bc = Γ a
cb − Γ a
bc,
with
Ta
bc = Πi
b(Π
a
c ),i −Πi
c(Π
a
b ),i + Πa
i (Π
i
b),c −Πa
i (Π
i
c),b =
(
N[gΠ
)a
bc
. (4)
The final formulae for Γ ’s are given as:
Γ a
bc = 1
2gai
{
gib,c + gic,b − gbc,i
}
− 1
2gai
{
gijT
j
bc + gbjT
j
ci + gcjT
j
bi
}
=
{ a
b c
}
RLC
−Ka
bc.
The dynamics of a given state (p = position, v = velocity) at time t is governed by the
infinitesimal flow of the Lagrange dynamical vector field (2) evaluated at (p , v) ∈ TM . Let us
stress that its horizontal part is now specified by the modified set of Christoffel symbols. Apart
from the standard (commutative) RLC-term there appears the additional tensorial part Ka
bc
called the contortion. Thus the NC-Lagrange equations can be written as follows:
ẍa + ẋcẋb
{ a
b c
}
RLC
= F a + ẋcẋbKa
bc, a = 1, . . . ,m. (5)
Using the optics of the standard Lagrange mechanics: the “noncommutativity” can be treated
as an additional “internal” force. Its origin comes from the presence of the nonzero contortion.
To reveal its physical consequences let us recall what geometrically the contortion is responsible
for. Parallel transport (1) along γ determined by Γ a
bc =
{
a
b c
}
RLC
−Ka
bc can be split into two
steps:
va 7−→ va
r = va + εvb
[
Ka
bcγ̇
c
]
+ o(ε) 7−→ va
r − εγ̇cvb
r
{ a
b c
}
RLC
+ o(ε).
The first arrow is an infinitesimal linear transformation determined by K(·, γ̇) that takes place
in the same tangent space as the initial vector v. The second one is just the standard Levi-
Civita parallel transport of vr. Since the considered connection is compatible with the metric,
both v and vr have the same lengths and therefore K(·, γ̇) is a generator of some rotation (it
depends on the tangent vector γ̇ to the transporting curve γ). Thus the effect of additional
noncommutative “Lorentz-like” force is an infinitesimal rotation of actual velocity that precedes
the next “commutative” RLC-step. Let us recall that mechanics with non zero torsion is the
subject of study at [12], similar problem from the general relativity point of view is analyzed
in [13].
Example. Suppose that M is ordinary two-dimensional plane, which is endowed by the flat
metric g = 1
2 {dx ⊗ dx + dy ⊗ dy}. Noncommutativity on M is imposed by a general Poisson
bracket:
{x, y} = Θ(x, y), here Θ(x, y) is any (well-behaved) function on R2.
The dynamics of a free particle is then governed by the NC-force, whose Cartesian components
are given as follows:
F x
NC
= ẏΘ{ẋΘ,y − ẏΘ,x}, and F y
NC
= −ẋΘ{ẋΘ,y − ẏΘ,x}.
Noncommutative Lagrange Mechanics 7
4 Concluding remarks
• The dynamical system we have considered was an autonomous one. But in a general metric,
noncommutativity, as well as the external strengths, could be explicitly time dependent. In
that case, one should work with the connection on the first jet-extension of M × R. Emerging
complications are mainly technical.
• The Lorentz-like force specifying the coupling of the noncommutative background with the sys-
tem under consideration comes from pure geometrical considerations. Very similar magnetic-like
background effect of the noncommutativity emerges also in the standard Moyal plane approach,
see [5, 6, 7]. Here it is due to the affine connection with the nonzero contortion. There it is so,
because
[xa, xb] = iΘab (a, b = 1, 2)
in two-dimensions provides an effective coupling with the axial magnetic field B3 = Θ12. Both
approaches to the noncommutative mechanics are therefore ideologically independent and phys-
ically inequivalent. This can be easily seen for example, if Θab = const εab. Then there is
a nonzero magnetic field B = (0, 0,B3), but since Kc
ab depends on derivatives of Θab, the con-
tortion term is zero.
• The question of symmetries of equations of motion has an obvious solution. The base vector
field V = V i(x)∂xi ∈ Γ(TM) is a generator of symmetry of the considered dynamical system
specified by (5), if its complete (natural, also called horizontal) tangent bundle lift
V c = V a(x)∂xa + ẋb
(
V a(x)
)
,b
∂ẋa ,
commutes with the Lagrange dynamical field L. Let me remind the reader that L and V c are
both the vector fields defined on the tangent bundle TM .
• The additional contortion dependent force emerging in (5) is not potential-generated nei-
ther generalized potential-generated, i.e. there does not exist a tangent bundle potential func-
tion U(x, ẋ), such that
Kabcẋ
bẋc ≡ gaiKi
bcẋ
bẋc = − ∂U
∂xa
+
d
dt
(
∂U
∂ẋa
)
.
The proposed NC-dynamics is therefore not derivable from some “noncommutative” Lagrangian
and/or Hamiltonian function in the obvious fashion. The question about the possible quantiza-
tion thus remains open (at least for the author of this paper). Since we do not have Hamilto-
nian H, we can not write:
ẋ = {x,H}?~
, ṗ = {p, H}?~
, where {f,h}?~
:= f ?~ h− h ?~ f
and the Moyal–Kontsevich ?~-product is defined with respect to the given Poisson structure Π
which specifies the noncommutative background under consideration.
• Suppose that the external forces are potential-generated, then the classical energy can be
introduced as standardly:
E = 1
2gab(x)ẋaẋb + U(x) = T + U.
Verification that E is conserved by the NC-dynamics is straightforward2.
2Concisely, take the scalar product of (5) with the actual velocity ẋa and use the mentioned “Lorentz-likeness”
of the additional contortion term ẋaKabcẋ
bẋc = 0 = g(v,K(v, v)), then
gabẋ
a
[
ẍb + ẋiẋj
{ b
j i
}
RLC
]
= gabẋ
aẍb + 1
2
gab,iẋ
aẋbẋi(= Ṫ
) (5)
= −ẋaU,a + 0 = −U̇ =⇒ Ė = 0.
8 D. Kochan
• The dynamics in the Hamiltonian “picture” is just a simple combination of cotangent bundle
parallel transport3 and vertical lift of the co-vectorial strength field Q. Suppose that at some
time t the mechanical system occupies a cotangent bundle state (p = position, α = momentum),
then at infinitesimally close time the new T ∗M -position is determined by the flow of the Hamil-
ton dynamical field:
H
∣∣∣
(p,α)
=
(
]gα
)∣∣∣Hor
(p,α)
+ Q
∣∣∣Ver
(p,α)
∈ T(p,α)(T
∗M).
• The configuration (Riemannian) space (M, g) could be at any time replaced by a pseudo-
Riemannian space-time manifold. Then one can easily impose “noncommutativity” within gene-
ral relativity modifying the underlaying metric connection by non zero torsion terms. General
relativity with torsion is analyzed at [13].
Acknowledgements
This research was supported in part by MŠMT ČR grant LC06002, ESF projects: JPD 3
2005/NP1-013 and JPD 3BA 2005/1-034 and VEGA Grant 1/3042/06. The author is thankful
to Pavel Exner, Miloslav Znojil and Jaroslav Dittrich for their kindness and hospitality during
the author’s short-time fellowship at the Doppler Institute in the autumn 2006.
References
[1] Seiberg N., Witten E., String theory and noncommutative geometry, J. High Energy Phys. 1999 (1999),
no. 9, 032, 93 pages, hep-th/9908142.
[2] Doplicher S., Fredenhagen K., Roberts J.E., The quantum structure of spacetime at the Planck scale and
quantum fields, Comm. Math. Phys. 172 (1995), 187–220, hep-th/0303037.
[3] Connes A., Noncommutative geometry, Academic Press, 1994.
Madore J., An introduction to noncommutative differential geometry and its physical applications, Cam-
bridge University Press, Cambridge, 1999.
Gracia-Bondia J.M., Várilly J.C., Figueroa H., Elements of noncommutative geometry, Birkhäuser, Boston,
2001.
[4] Douglas M.R., Nekrasov N.A., Noncommutative field theory, Rev. Modern Phys. 73 (2001), 977–1029,
hep-th/0106048.
Szabo R.J., Quantum field theory on noncommutative spaces, Phys. Rep. 378 (2003), 207–299,
hep-th/0109162.
[5] Nair V.P., Polychronakos A.P., Quantum mechanics on the noncommutative plane and sphere, Phys. Lett. B
505 (2001), 267–274, hep-th/0011172.
Gamboa J., Loewe M., Rojas J.C., Noncommutative quantum mechanics, Phys. Rev. D 64 (2001), 067901,
3 pages, hep-th/0010220.
Demetrian M., Kochan D., Quantum mechanics on non-commutative plane, Acta Phys. Slov. 52 (2002),
1–9, hep-th/0102050.
Djemai A.E.F., Smail H., On quantum mechanics on noncommutative quantum phase space, Commun.
Theor. Phys. (Beijing) 41 (2004), 837–844, hep-th/0309006.
Ho P.M., Kao H.C., Noncommutative quantum mechanics from noncommutative quantum field theory,
Phys. Rev. Lett. 88 (2002), 151602, 4 pages, hep-th/0110191.
Dragovich B., Rakic Z., path integrals in noncommutative quantum mechanics, Theoret. and Math. Phys.
140 (2004), 1299–1308, hep-th/0309204.
[6] Duval C., Horváthy P.A., The exotic Galilei group and the “Peierls substitution”, Phys. Lett. B 479 (2000),
284–290, hep-th/0002233.
3Concisely, parallel transport of a co-vector α ∈ T ∗
p M along γ is effectively defined via its g-dual, i.e. the
vector ]gα ∈ TpM . Transporting it and taking its inverse dual at a second terminal point of γ one gets parallel
transport of the co-vector α.
http://arxiv.org/abs/hep-th/9908142
http://arxiv.org/abs/hep-th/0303037
http://arxiv.org/abs/hep-th/0106048
http://arxiv.org/abs/hep-th/0109162
http://arxiv.org/abs/hep-th/0011172
http://arxiv.org/abs/hep-th/0010220
http://arxiv.org/abs/hep-th/0102050
http://arxiv.org/abs/hep-th/0309006
http://arxiv.org/abs/hep-th/0110191
http://arxiv.org/abs/hep-th/0309204
http://arxiv.org/abs/hep-th/0002233
Noncommutative Lagrange Mechanics 9
Horváthy P.A., Plyushchay M.S., Nonrelativistic anyons, noncommutative plane and exotic Galilean sym-
metry, J. High Energy Phys. 2002 (2002), no. 6, 033, 11 pages, hep-th/0201228.
Banerjee R., A novel approach to noncommutativity in planar quantum mechanics, Modern Phys. Lett. A
17 (2002), 631–645, hep-th/0106280.
Horváthy P.A., Plyushchay M.S., Anyon wave equations and the noncommutative plane, Phys. Lett. B 595
(2004), 547–555, hep-th/0404137.
[7] Romero J.M., Santiago J.A., Vergara J.D., Newton’s second law in a non-commutative space, Phys. Lett. A
310 (2003), 9–12, hep-th/0211165.
Djemai A.E.F., On noncommutative classical mechanics, Internat. J. Theoret. Phys. 42 (2004), 299–314,
hep-th/0309034.
[8] Arnol’d V.I., Mathematical methods of classical mechanics, 2nd ed., Springer-Verlag, New York, 1989.
Abraham R., Marsden J.E., Foundation of mechanics, Addison-Wesley, 1978.
Burke W.L., Applied differential geometry, Cambridge University Press, 1985.
Fecko M., Differential geometry and Lie groups for physicists, Cambridge University Press, 2006.
[9] Kobayashi S., Nomizu K., Foundation of differential geometry, Vol. I, Wiley, New York, 1963.
do Carmo M.P., Riemannian geometry, Birkhäuser, Boston, 1992.
Frankel T., The geometry of physics, Cambridege University Press, Cambridge 1997.
[10] Bayen F., Flato M., Frønsdal C., Lichnerowicz A., Sternheimer D., Deformation theory and quantization,
Ann. Phys. 111 (1978), 61–151.
Fedosov B.V., A Simple Geometrical construction of deformation quantization, J. Differential Geom. 40
(1994), 213–238.
Kontsevich M., Deformation quantization of Poisson manifolds, Lett. Math. Phys. 66 (2003), 157–216,
q-alg/9709040.
[11] Frölicher A., Nijenhuis A., Theory of vector-valued differential forms. I. Derivations of the graded ring of
differential forms, Nederl. Akad. Wetensch. Proc. Ser. A 59 (1956), 338–359.
[12] Shabanov S.V., Constrained systems and analytical mechanics in spases with torsion, J. Phys. A: Math.
Gen. 31 (1998), 5177–5190, physics/9801023.
[13] Hehl F.W., von der Heyde P., Kerlick G.D., Nester J.M., General relativity with spin and torsion: founda-
tions and prospects, Rev. Modern Phys. 48 (1976), 393–416.
Hehl F.W., McCrea J.D., Mielke E.W., Ne’eman Y., Metric-affine gauge theory of gravity: field equations,
Noether identities, world spinors, and breaking of dilation invariance, Phys. Rep. 258 (1995), 1–171.
http://arxiv.org/abs/hep-th/0201228
http://arxiv.org/abs/hep-th/0106280
http://arxiv.org/abs/hep-th/0404137
http://arxiv.org/abs/hep-th/0211165
http://arxiv.org/abs/hep-th/0309034
http://arxiv.org/abs/q-alg/9709040
http://arxiv.org/abs/physics/9801023
1 Introduction and motivation
2 Preliminaries
3 Noncommutative Lagrange mechanics
4 Concluding remarks
References
|