A Simple Model of Double Dynamics on Lie Groups
Extracted source text ยท arXiv 2105.08638
%\documentclass[11pt,english]{article}
\documentclass[graybox]{svmult}
%%%%%%%% commands by journal %%%%%%%%%%%%%%%%%%
\usepackage{mathptmx} % selects Times Roman as basic font
\usepackage{helvet} % selects Helvetica as sans-serif font
\usepackage{courier} % selects Courier as typewriter font
\usepackage{type1cm} % activate if the above 3 fonts are
% not available on your system
%
\usepackage{makeidx} % allows index generation
\usepackage{graphicx} % standard LaTeX graphics tool
% when including figure files
\usepackage{multicol} % used for the two-column index
\usepackage[bottom]{footmisc}% places footnotes at page bottom
% see the list of further useful packages
% in the Reference Guide
\makeindex % used for the subject index
% please use the style svind.ist with
% your makeindex program
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\usepackage{graphicx}
%\usepackage{physics}
\usepackage{amstext}
\usepackage{amsbsy}
%\usepackage{calrsfs}
%\usepackage{unicode-math}
%\setmathfont{XITS Math}
%\setmathfont[version=setB,StylisticSet=1]{XITS Math}
%\usepackage{caption}
\usepackage{etoolbox}
%\usepackage{makeidx}
\include{epsf}
\usepackage{amsfonts}
\usepackage{authblk}
\usepackage{sectsty}
\usepackage{amsmath,amssymb,epsfig}
\usepackage{amscd}
%\usepackage{amsthm}
\usepackage{mathrsfs}
\usepackage{dsfont}
\usepackage[applemac]{inputenc}
\usepackage[english]{babel}
\usepackage{enumitem}
\usepackage[]{latexsym}
\usepackage{braket}
%\usepackage{caption}
%\usepackage{hyperref}
%\hypersetup{
%pdftitle={},%
%pdfauthor={},%
%pdfsubject={},%
%pdfkeywords={},%
%colorlinks=true,%
%linkcolor=blue,%
%citecolor=red,%
%linktocpage=true,%
%hyperfootnotes=true,%
%pageanchor=true
%}
%%%%%%%%%%%%%%%% i miei comandi (patrizia) %%%%%%%%%%
\renewcommand{\theequation}{\thesection.\arabic{equation}}
\newcommand{\be}{\begin{equation}}
\newcommand{\ee}{\end{equation}}
\def\beqa{\begin{eqnarray}}
\def\eeqa{\end{eqnarray}}
\def\bean{\begin{eqnarray*}}
\def\eean{\end{eqnarray*}}
\def\nn{ \nonumber}
\def\dd{{\bf d}}
\newcommand{\Rl}{\mathbb{R}^3_\lambda}
\newcommand{\Rt}{\mathbb{R}^4_\theta}
\newcommand{\R}{\mathbb{R}}
\newcommand{\C}{\mathbb{C}}
\newcommand{\N}{\mathbb{N}}
\newcommand{\tg}{{\tilde{g}}}
\newcommand{\te}{{\tilde{e}}}
\newcommand{\tQ}{{\tilde{Q}}}
\newcommand{\tI}{{\tilde{I}}}
\newcommand{\tJ}{{\tilde{J}}}
\newcommand{\tX}{{\tilde{X}}}
\newcommand{\tY}{{\tilde{Y}}}
\newcommand\gl{\mathfrak}
\newcommand\dl{\mathbb}
\newcommand{\ad}{\hbox{ad}}
%\renewcommand{\thefootnote}{\fnsymbol{footnote}}
\newcommand{\eqn}[1]{(\ref{#1})}
\newcommand{\del}{\partial}
\newcommand{\Tr}[1]{\:{\rm Tr}\,#1}
%\newcomTmand{\dd}{{\mathrm d}}
\def\baselinestretch{1.1}
\textheight 23.5cm \textwidth 16cm
\parskip 1ex
\oddsidemargin 0pt \evensidemargin 0pt \topmargin -60pt \jot = .5ex
\parskip 1ex
\renewenvironment{thebibliography}[1]
{\section*{References}\frenchspacing\small
\begin{list}{[\arabic{enumi}]}
{\usecounter{enumi}\parsep=2pt\topsep 0pt
\settowidth{\labelwidth}{[#1]}
\leftmargin=\labelwidth\advance\leftmargin\labelsep
\rightmargin=0pt\itemsep=1pt\sloppy}}{\end{list}}
% \numberwithin{equation}{section}
\begin{document}
\title*{A simple model of double dynamics on Lie groups}
%\\
%\textcolor{red}{\it oppure}\\
%A toy model for Double Field Theory and its generalization to Principal Chiral Models\\
%\textcolor{red}{\it oppure}\\
%Lie-Poisson duality for the Rigid Rotator and its Double Field Theory Generalization
%\vspace{0.5cm}}
%\date{}
%\author[1,3]{Vincenzo E. Marotta}
%\author[2]{Franco Pezzella}
\author{Patrizia Vitale}
%\\
\institute{Patrizia Vitale \at Dipartimento di Fisica ``E. Pancini'', Universit\`a di Napoli Federico II
and
INFN-Sezione di Napoli,
Complesso Universitario di Monte S. Angelo Edificio 6, via Cintia, 80126 Napoli, Italy. \email{patrizia.vitale@na.infn.it}}
%\affil[3]{\textit{\footnotesize Department of Mathematics,
%Heriot-Watt University
%Colin Maclaurin Building, Riccarton, Edinburgh EH14 4AS, U.K.}}
%\affil[ ]{}
%\affil[ ]{\footnotesize e-mail: \texttt{patrizia.vitale@na.infn.it}}
%\begin{document}
\maketitle
\abstract*{
We study the dynamics of the rigid rotator on the group manifold of $SU(2)$ as an instance of dynamics on Lie groups, together with that of a model whose carrier space is the Borel group $SB(2,\C)$, the Lie Poisson dual of $SU(2)$. We thus introduce a parent action on the Drinfel'd double of the above mentioned groups, which describes the dynamics of a system with twice as many degrees of freedom as the two starting partners. Through a gauging procedure of its global symmetries both the rigid rotor and the dual model are recovered.
\\
~
\\
{\it keywords: Generalized Geometry, Double Field Theory, T-Duality, Poisson-Lie symmetry}
}
\abstract{
We study the dynamics of the rigid rotator on the group manifold of $SU(2)$ as an instance of dynamics on Lie groups, together with that of a model whose carrier space is the Borel group $SB(2,\C)$, the Lie Poisson dual of $SU(2)$. We thus introduce a parent action on the Drinfel'd double of the above mentioned groups, which describes the dynamics of a system with twice as many degrees of freedom as the two starting partners. Through a gauging procedure of its global symmetries both the rigid rotor and the dual model are recovered.
\\
~
\\
{\it keywords: Generalized Geometry, Double Field Theory, T-Duality, Poisson-Lie symmetry}
}
%\newpage
%\tableofcontents
\section{Introduction}
This paper is based on a lecture given at the conference in honour of Alberto Ibort, "Classical and Quantum Physics: Geometry, Dynamics and Control" at ICMAT, Madrid, in March 2018, and it is aimed at discussing within a simple example of dynamics over Lie groups, the isotropic rigid rotator (IRR), the interplay between concepts such as non-Abelian and Poisson-Lie T-duality \cite{Klim, Klim2}, Generalized and Doubled Geometry \cite{hitchin1, hitchin2, gualtieri:tesi, hulled}, Double Field Theory (DFT) \cite{HZ}, in the mathematical framework of Poisson-Lie groups and Drinfel'd doubles \cite{drinfel'd, semenov}. The main goal being here to convey the general philosophy, many technical details and in deep calculations are left aside and we refer to \cite{MPV18} for an extended presentation of the results. A generalization to field theory is in preparation and will appear hopefully soon \cite{MPV18_2}.
The cotangent space of a $d$ dimensional Lie group, $G$, $T^* G\sim G\times \R^d$ , while providing the carrier space for the Hamiltonian dynamics of many systems of physical relevance, possesses a very interesting structure from the mathematical point of view, it being itself a Lie group, the semi-direct product of the Lie group $G$ with the dual of its Lie algebra, the Abelian Lie algebra $\mathfrak{g}^*\sim\R^d$, thought of as an Abelian vector group. Free dynamics over the group manifold is described in terms of momenta, $I_i, i=1,...,d$ which are coordinate variables for the fiber $\mathfrak{g}^*$, with Hamiltonian evolution governed by Kirillov-Souriau-Konstant (KSK) Poisson brackets $\{I_i, I_j\}= c_{ij}^k I_k$ and $c_{ij}^k$ the structure constants of the Lie algebra $\mathfrak{g}$.
The first interesting, although well known, remark is thus that, as a Lie group $\mathfrak{g}^*$ is Abelian, but the Poisson algebra of linear functions over $\mathfrak{g}^*$, is non-Abelian and isomorphic to the Lie algebra $\mathfrak{g}$. Moreover the latter can be obtained by linearizing the so-called Poisson-Lie structure over the dual group of $G$ which shall be indicated by $G^*$\cite{marmo:articolo1}.
The tangent bundle $TG$ has the structure of a group, with its fiber which is isomorphic to the Lie algebra $\mathfrak{g}$, typically non-Abelian. Fiber coordinates, namely the generalized velocities, are components of left or right invariant vector fields $X_a = \dot Q^i_{(a)} \frac{\del}{\del Q^i}, a=1,...,d$ with non-trivial Lie brackets $[X_a, X_b]= c_{ab}^c X_c $.
Models which exchange tangent and cotangent bundle coordinates over Lie groups, have been widely studied in the context of field theory of sigma models and the duality which is naturally inherited from the relation between the Lie algebra and its dual, is referred to as non-Abelian, or semi-Abelian duality \cite{Klim}. According to the structure which we have outlined, the dual model in such context relies on a Poisson algebra of momenta which is Abelian. The IRR system, which we are going to investigate in the paper, can be regarded as a $0+1$ dimensional analogue of a non-linear sigma model with target space $SU(2)$.
Once identified the cotangent space of the Lie group $G$ with a semi-direct product of groups, the second interesting observation is that there exists a well defined procedure for deforming the semi-direct product into a fully non-Abelian group, by introducing a non-trivial Lie algebra structure over $\mathfrak{g}^*$, so that the cotangent space $G\ltimes\R^d$ be replaced by $D\simeq G\cdot G^*$ (the latter trivialization being only local) with $G^*$ the Lie group obtained by $\mathfrak{g}^*$ by exponentiation. The group $D$ is the Drinfel'd double of $G$. Since the role of $G$ and $G^*$ is symmetric, $D$ can be dually regarded as the double of $G^*$ and hence, as a deformation of the cotangent bundle $ T^*G^*$.\footnote{Notice however that, in case $G$ is a compact group, such as $SU(2)$, its cotangent bundle is truly diffeomorphic to its Drinfel'd double, while the cotangent bundle of its dual, $SB(2\C)$ for $SU(2)$, is only a deformation of the double.} Therefore a dynamical system can be defined with the latter as carrier space of the dynamics. The new algebra of momenta, being dictated by the KSK Poisson brackets $\{\tilde I^i,\tilde I^j\}= f^{ij}_k \tilde I^k $ ($f^{ij}_k$ the structure constants of $G^*$), can be retrieved by linearizing the Poisson-Lie bracket over $G$ \cite{MPV18}. Therefore, the IRR model and the latter one are dual to each other, in the sense that they are defined over partner groups in the Drinfel'd double $D$, with linearized Poisson-Lie brackets of momenta. This is a kind of Poisson-Lie duality, although the name has already been used in the literature in the context of sigma models, implying a more stringent relation, which entails the metrics of the two models as well \cite{Klim2}. Completely integrable systems and their relation to double Lie groups have been studied in \cite{MI98}.
A natural step further is to consider a parent action over the Drinfel'd double $D$ with twice as many degrees of freedom as the two actions over $G$, $G^*$ respectively. Such a doubling will generate a Double Field Theory when considering the 1+1-dimensional analogue of the IRR model \cite{MPV18_2}. The generalized action encodes the global symmetries of both models and can be reduced to either of them by means of an appropriate gauging of some of the symmetries. The generalized momenta ${\bf P_I}$ being sections of the cotangent bundle $T^*D$ can be described in terms of the momenta of the cotangent spaces of the dually related groups $G, G^*$: ${\bf P_I}\simeq (I_i, \tilde I^i)$, where the position of indices stresses the dual geometric meaning of the two. Because the cotangent fibre of one group can be identified with the tangent fiber of the other and vice-versa, the connection with Generalized Geometry is thus natural.
The latter was first introduced by N. J. Hitchin in Ref. \cite{hitchin1}. In a nutshell, it consists in replacing the tangent bundle $T$ of a manifold $M$ with $T\oplus T^*$, a bundle with the same base space $M$ but fibers given by the direct sum of tangent and cotangent spaces. Moreover, the Lie bracket on the sections of $T$, which are vector fields, is replaced by the Courant bracket which involves vector fields and one-forms. This is in turn related to Double Field Theory (DFT) \cite{HZ}.
DFT was introduced to realize the symmetry of the dynamics of string theory under T-duality transformations as a manifest symmetry of the string action. In order to achieve this goal, the degrees of freedom of the target space, represented by the coordinate fields $x^i(\sigma,\tau)$, $i = 1, \dots, d$ have to be {\em doubled} with respect to the original. Therefore, in the framework of string theory, the doubling takes place in the $d$-dimensional target space $M$ of the non-linear sigma model underlying the string action, by introducing new fields $\tilde x_i (\sigma,\tau)$, which are dual to $x^i(\sigma,\tau)$, with $i = 1, \dots, d$. This is in perfect agreement with the approach of Generalized Geometry, by identifying $x^i, \tilde x_i $ with sections of a generalized bundle $E\oplus E^*$ over the world sheet of the string. Thus, it is only when the target space of the string becomes the configuration space of an {\it effective field theory}, that the doubling of fields is reinterpreted as a doubling of the configuration space. A formulation of the string action with manifest T-duality invariance was first proposed by \cite{Tseytlin, Duff}. A corresponding doubling of the space-time degrees of freedom in the low-energy effective action first occurred in the pioneering of work of Siegel \cite{siegel}.
The interest for DFT is naturally not limited to string theory but it
is relevant in the broad area of field theory when one deals with duality symmetries of the dynamics which are not manifest at the level of the action. For an early paper see for example \cite{siegel84}, which addresses the problem of constructing manifestly Lorentz invariant actions for self dual gauge fields by means of the introduction of auxiliary fields in the Lagrangian.
%When considering vector fields and one-forms as components of a generalized (indeed doubled) vector field on the carrier space of the effective dynamics (which is itself doubled), one has, on the one hand, another instance of field doubling, but on the other hand, at the same time, a section of a generalized tangent bundle as in Generalized Geometry.
% The geometry of the effective dynamics is thus more appropriately renamed {\em Doubled Geometry} (DG).
To summarize, doubling can occur at different stages:
\begin{itemize}
\item at the level of fields on a given configuration space, as in GG;
\item at the level of configuration space coordinates, with fields $\phi$ depending on twice the initial configuration space variables, $\phi= \phi(x^i, \tilde x_i)$, as for the strings effective dynamics. In this case it is more appropriate to talk about {\em Doubled Geometry} (DG) \cite{Hull, hulled};
\item at the level of both, fields and coordinates as for example in DFT of the strings effective dynamics, when dealing with the symmetries of the generalized action.
\end{itemize}
There is therefore an interplay between GG and DG on the one hand and DFT on the other hand which, within the framework we have sketched, emerges from the identification of the appropriate carrier space of the dynamics.
It is then clear that models whose dynamics takes place on Lie groups can be very helpful in better understanding the above mentioned relations, because the notion of Drinfel'd doubles and Poisson-Lie dual groups, together with their symmetries, is well established \cite{drinfel'd, semenov}. The idea of investigating such geometric structures in relation to duality in field theory has already been applied to sigma models by Klim\v{c}\'ik and \v{S}evera in \cite{Klim, Klim2, lledo} where the authors first introduced the notion of Poisson-Lie T-duality (also see \cite{sfetsos}, \cite{falceto}). DFT on group manifolds with its relation with Poisson-Lie symmetries, has been studied in \cite{hassler}.
Since our model describes an example of particle dynamics, the most appropriate doubling within those enumerated above is the doubling of the configuration space. For the same reason, we shall see that the model considered here is too simple to exhibit symmetry under duality transformation, although a generalization to field theory is possible \cite{MPV18_2}.
The paper is intended to give a short account of the results contained in \cite{MPV18}. Therefore the structure strictly follows the former, although being much shorter and less technical. In Sect. \ref{rigidrot} we review the dynamics of the IRR on the group manifold of the group $SU(2)$. In Sect. \ref{drinfel'd} we shortly introduce the mathematical framework of Poisson-Lie groups and Drinfel'd doubles. In Sect. \ref{dualrot} we introduce a dynamical model on the dual group of $SU(2)$, the Borel group $SB(2,\C)$ and study its dynamics.
In Sect. \ref{gensec} we present the parent action on the Drinfel'd double, which has twice as many degrees of freedom as the two partner models. The latter are recovered by appropriately gauging the global symmetries of the former. Finally, in Sect. \ref{hamform} we introduce the Hamiltonian formalism for the double model and in \ref{canform} we study in detail the full Poisson algebra, together with the Hamiltonian vector fields associated with momenta $(I_i,\tI^i)$. The Poisson brackets of the generalized momenta $(I_i,\tI^i)$ can be related to Poisson-Lie brackets on the two dual groups. The full Poisson algebra of momenta is isomorphic to the algebra of $SL(2,\C)$, namely a semisimple group, with each set of momenta underlying a non-Abelian algebra. That is why we refer to the two models as non-Abelian duals giving rise to a kind of Poisson-Lie T-duality.
%Moreover, the latter can be interpreted as tangent and cotangent space coordinates on the generalized $T\oplus T^*$ bundle over $SU(2)$ (or, dually, over the group $SB(2,\C)$), with Poisson brackets yielding a Poisson realization of C-brackets.
Sect. \ref{concl} contains our conclusions.
\section{The Isotropic Rigid Rotator}\label{rigidrot}
The Isotropic Rigid Rotator (IRR) provides a classical example of dynamics over Lie groups (see \cite{marmosaletan} as a reference text for the subject). The tangent [cotangent] bundle of the group $SU(2)$ can be chosen as carrier space for the dynamics in the Lagrangian [Hamiltonian] formulation.
A suitable action for the system is thus the following
\be
S_0= \int_\R L _0~dt=-\frac{1}{4} \int_\R \Tr ( \phi^{*}(g^{-1} d g) \wedge *_H \phi^*(g^{-1} dg)) =-\frac{1}{4}\int_\R \Tr (g^{-1}{\dot g})^2 dt \label{lag}
\ee
where $\phi :t\in \R\rightarrow SU(2)$ is the group-valued dynamical variable, while $\phi^*$ is the induced pull-back map. $
g^{-1} d g = 2 i \alpha^k e_k $ is the Maurer-Cartan left-invariant one-form, $e_k=\sigma_k/2 $ are the $SU(2)$ generators with $\sigma_k$ the Pauli matrices, $\alpha^k$ are the basic left-invariant one-forms, $*_H$ denotes the Hodge
star operator on the source space $\R$, such that $*_H dt = 1$, and $\Tr$ the trace over the Lie algebra. Moreover,
\be g^{-1}{\dot g}\equiv \phi^*(g^{-1}{ d g}) (\Gamma)
\ee
with $\Gamma=d/dt$ . The model can be regarded as a $(0+1)$-dimensional field theory which is group-valued.
From now on, with an abuse of notation, we shall trade $g^{-1} d g$ for its pull-back, and the dynamical variable $\phi$ for $g$, as it is customary in the dynamics of fields over Lie groups. The group manifold can be parametrized with $\R^4$ coordinates as $g= 2(y^0 e_0
+i y^i {e_i})$, with $(y^0)^2+ \sum_i (y^i)^2=1$, $e_0={\mathbb I}/2$.
%One has then:
%\be
%y^0= \Tr (g e_0), \;\;\; y^i=-{i} \Tr (g e_i) \; \; \; i=1, \dots ,3
%\ee
By observing that
\be
g^{-1} \dot g = 2i (y^0 \dot y^i-y^i \dot y^0+ {\epsilon^{i}}_{jk} {y^j \dot y^k })e_i \label{qdot}
\ee
we define the left generalized velocities $\dot Q^i$ as
\be
\dot Q^i \equiv
(y^0\dot y^i-y^i \dot y^0+ {\epsilon^{i}}_{jk} y^j \dot y^k) \label{genvel}
\ee
so that $g^{-1} \dot g=2 i \dot{Q}^{i} e_{i}$.
$(Q^i, \dot Q^i)$ $i=1, \dots ,3$ are therefore tangent bundle coordinates, with $Q^i$ implicitly defined. From right-invariant one-forms one could define right generalized velocities in an analogous way. They give an alternative set of coordinates over the tangent bundle.
The Lagrangian $L_{0}$ in Eq. \eqn{lag} can be rewritten as:
\be
L_0=\frac{1}{2} \dot Q^i \dot Q^j \delta_{i j}.
\ee
In the intrinsic formulation the Euler-Lagrange equations of motion can be restated as
\be
{\sf L}_\Gamma \theta_{L} -d L_0=0
\ee
with
\be
\theta_L= \frac{1}{2}\Tr[ g^{-1} \dot g \;g^{-1} d g]= \dot Q^i \alpha^j \delta_{i j }
\ee
the Lagrangian one-form and ${\sf L}_{\Gamma}$ the Lie derivative with respect to $\Gamma$. With a little calculation we obtain
%By projecting along the basic left-invariant vector fields $X_i$ dual to $\alpha^i$, one obtains:
%\be
%i_{X_i} [ {\sf L}_\Gamma \theta_{L} -d L_0]=0
%\ee
%Since $ {\sf L}_{\Gamma}$ and $i_{X_i}$ commute over the Lagrangian one-form, one gets:
%\be
%{\sf L}_\Gamma(\dot Q^j\, i_{X_i}\alpha^l )\delta_{jl} - {\sf L}_{X_i} L_0 = 0
%\ee
%which implies
\be
{\sf L}_\Gamma\dot Q^j \delta_{ji} - \dot Q^p \dot Q^q {\epsilon_{ip}}^k\delta_{qk} = {\sf L}_\Gamma\dot Q^j \delta_{ji}= 0 \label{eqmo}
\ee
because of the rotation invariance of the product and the complete antisymmetry of the structure constants of $SU(2)$.
%
%Equivalently, the equations of motion can be rewritten as:
%\be
%\frac{d}{dt} \left(g^{-1}\frac{dg}{dt}\right)=0\label{eom}
%\ee
%being, from Eq. \eqn{qdot}:
%\be
%\delta_{ij} \dot{Q}^j= -{i}\Tr( g^{-1} \dot g \; e_i) .
%\ee
Cotangent bundle coordinates can be chosen to be $(Q^i, I_i)$ with the $I_i$'s denoting the left momenta $
I_i= \frac{\del L_0}{\del \dot Q^i}= \delta_{i j} \dot Q^j $. An alternative set of fiber coordinates is represented by the right momenta, which are defined in terms of the right generalized velocities.
The Legendre transform from $TSU(2)$ to $T^*SU(2)$ yields the Hamiltonian function:
\be
H_0=[I_i \dot Q^i -L_{0}]_{\dot Q^i=\delta^{ij} I_j }= \frac{1}{2} \delta^{ij}I_i I_j \label{h0} \,\,.
\ee
By introducing a dual basis $\{{e^i}^*\}$ in the cotangent space, such that $\langle{e^i}^*|e_j\rangle =\delta^i_j$, one can consider the linear combination:
\be
I=i\; I_i {e^i}^*. \label{Iform}
\ee
The dynamics of the IRR is thus obtained from the Hamiltonian \eqn{h0} and the following Poisson brackets
\beqa
\{y^i,y^j\}&=&0\label{pp}\\
\{I_i,I_j\}&=& {\epsilon_{ij\;}}^k I_k \label{xx}\\
\{y^i,I_j\}&=&\delta^{i}_j y^0 +{\epsilon^i}_{\; jk}y^k \;\;\; {\rm or ~ equivalently}\;\;\; \{g, I_j\}= 2 i g e_j \label{ij}
\eeqa
which can be easily derived from the action \cite{MPV18}.
%
%\be
%S_1= \int \langle I | g^{-1}\dot g\rangle dt - \int H_{0} \, dt \equiv \int \vartheta -\int H_{0} dt \nonumber
%\ee
%with $\theta$ the canonical one-form.
%Indeed the symplectic form $\omega$ is readily obtained as
%\be
%\omega= d \vartheta= d I_i \wedge \delta^{i}_j \alpha ^j - \frac{1}{2}I_i \delta_j^i {\epsilon^j}_{kl} \alpha^k\wedge \alpha^l \nonumber
%\ee
%with $ d\alpha^k= \frac{i}{2}\,{\epsilon^{k}}_{ij}\alpha^i\wedge\alpha^j$. By inverting $\omega$ one finds the Poisson algebra \eqn{pp}-\eqn{ij}.
The fiber coordinates $I_i$ are associated to the angular momentum components and the base space coordinates $(y^0, y^i)$ to the orientation of the rotator. The resulting system is rotationally invariant since
$
\{I_i, H_0\} = 0. $
The Hamilton equations of motion for the system read
\be
\dot I_i= 0,\;\;\; g^{-1}\dot g= 2 i I_i \delta^{ij} e_j. \nonumber
\ee
Thus the angular momentum
$I_i$ is a constant of motion, while $g$ undergoes a uniform
precession. Since the Lagrangian and the Hamiltonian are invariant under right and left $SU(2)$ action, as well-known right momenta are conserved as well, being the model super-integrable.
Let us remark here that, while the fibers of the tangent bundle $TSU(2)$ can be identified, as a vector space, with the Lie algebra of $SU(2)$, $\mathfrak{su}(2)\simeq \R^3$, with $\dot Q^i$ denoting vector fields components, the fibers of the cotangent bundle $T^*SU(2)$ are isomorphic to the dual Lie algebra $\mathfrak{su}(2)^*$. As a vector space this is again $\R^3$, but the $I_i$ 's are now components of one-forms. This remark will be relevant in the next section, when the Abelian structure of $\mathfrak{su}(2)^*$ is deformed.
As a group, $T^*SU(2)$ is the semi-direct product of $SU(2)$ and the Abelian group $\mathbb{R}^3$, with the corresponding Lie algebra given by:
\be \label{JJ}
\left[L_i,L_j\right] =i {\epsilon_{ij}}^k L_k, ~~~~~ \left[T_i,T_j\right] = 0, ~~~~~\left[L_i,T_j\right] =i {\epsilon_{ij}}^k T_k.
\ee
Thus, the non-trivial Poisson algebra on the cotangent space reflects the non-triviality of the latter.
%Then, the non-trivial Poisson bracket on the fibers of the bundle, \eqn{xx}, can be
%understood in terms of the coadjoint action of the group $SU(2)$ on its dual algebra
%$\mathfrak{su}(2)^*\simeq \mathbb{R}^3$ and it reflects the
%non-triviality of the Lie bracket \eqn{JJ}. In this picture the Lie algebra generators $L_i$'s are
%identified with the linear functions on the dual algebra.
%\footnote{The group semi-direct product structure of the phase space $T^*SU(2)$ has been widely investigated in literature in many different contexts. Besides classical, well known applications, some of which we have already mentioned in the introduction, let us mention here applications in noncommutative geometry in relation to the quantization of the hydrogen atom \cite{duflo}, to the electron-monopole system \cite{noi}, and, recently, to models on three-dimensional space-time with $\mathfrak{su}(2)$ type non-commutativity \cite{vitalewallet, kupr, wallet, zoupanos}}.
%Before concluding this short review of the canonical formulation of the dynamics of the rigid rotator, let us stress the main points which we are going to elaborate further:
%\begin{itemize}
%\item The carrier space of the Hamiltonian dynamics is represented by the semi-direct product of a non-Abelian Lie group, $SU(2)$, and the Abelian group $\R^3$ which is nothing but the dual of its Lie algebra.
%\item The Poisson brackets governing the dynamics are the Kirillov-Poisson-Souriau brackets induced by the coadjoint action.
%\end{itemize}
Before closing the section let us mention that already in Ref. \cite{marmo:articolo1} the phase space of the rigid rotator was generalized to the semisimple group $SL(2, C)$, by replacing the Abelian subgroup $R^3$ of the semi-direct product above, with the Borel group $SB(2,\C)$, namely, passing to the {\it double} Lie group of $SU(2)$.
In next section we shall review the mathematical construction of Drinfel'd double Lie groups and their relation with the structures of Generalized Geometry. Our generalization will be however different form the one considered in \cite{marmo:articolo1}.
\section{Poisson-Lie Groups and the Double Lie Algebra $\mathfrak{sl}(2,\mathbb{C})$}\label{drinfel'd}
%In this section we shortly review Poisson-Lie groups and Drinfel'd doubles, see~\cite{semenov,alex,wein} for details, with the aim of introducing, in the forthcoming sections, new Lagrangian and Hamiltonian formulations of the IRR with a manifest symmetry under duality transformation. More precisely, in Section \ref{dualrot}, a model which is dual to the one described in Section \ref{rigidrot} is introduced, while in Section \ref{gensec} \textcolor{magenta}{ a new model is built} with doubled dynamical variables and with a manifest symmetry under duality transformation. The dynamics derived from the new action describes two models, dual to each other, being one the ordinary rigid rotator, the other a``rotator-like" system, with the rotation group $SU(2)$ replaced by its Poisson-Lie dual, the group $SB(2,C)$ of Borel $2\times 2$ complex matrices.
{ A
Poisson-Lie group \cite{semenov, alex, babelon, wein} is a Lie group equipped with a Poisson structure
which makes the product $\mu :G\times G \rightarrow G$ a Poisson map if
$G \times G$ is equipped with the product Poisson
structure. Linearization of the Poisson structure at the unit $e$ of
$G$ provides a Lie algebra structure over the dual algebra ${\gl g}^*=T^*_{e}(G)$ by the
relation
\begin{equation}
\label{Liedual}
[d\xi_{1}(e),d\xi_{2}(e)]^{*}=d\{\xi_{1},\xi_{2}\}(e)
\end{equation}
with $\xi_i\in C^\infty(G)$.
The compatibility condition between the Poisson and Lie structures of $G$ yields the relation:
\begin{equation}
\label{comp}
\left< [X,Y],[v,w]^*\right>+\left<\ad_v^*X,\ad_Y^*w\right>
-\left<\ad_w^*X,\ad_Y^*v\right>
-\left<\ad_v^*Y,\ad_X^*w\right>+\left<\ad_w^*Y,\ad_X^*v\right>=0\,
\end{equation}
with $v,w\in\mathfrak{g}^*, X,Y\in \mathfrak{g}$ and $ \ad_X^*, \ad_v^*$ the coadjoint actions of the Lie algebras $\mathfrak{g}, \mathfrak{g}^*$ on each other.
This allows one to define a Lie bracket in ${\gl g}\oplus {\gl g}^*$ through the formula:
\begin{equation}
\label{Liesuma}
[X+\xi,Y+\zeta]=[X,Y]+[\xi,\zeta]^{*}-ad^{*}_{X}\zeta + ad^{*}_{Y}\xi
+ ad^{*}_{\zeta}X - ad^{*}_{\xi}Y \,\, .
\end{equation}
%If $G$ is connected and simply connected,~(\ref{comp}) is enough to
%integrate $[\ ,\ ]^*$ to a Poisson structure on $G$ that makes it
%Poisson-Lie and the Poisson structure is unique.
The symmetry between
${\gl g}$ and ${\gl g}^*$ in~(\ref{comp}) implies that one has also a
Poisson-Lie group $G^*$ with Lie algebra $({\gl g}^*,[\ ,\ ]^*)$ and a
Poisson structure whose linearization at $e\in G^*$ gives the bracket $[\ ,\
]$. $G^*$ is the dual Poisson-Lie group of $G$.
%The two Poisson brackets on $G$, $G^*$, which are dually related to the Lie algebra structure on ${\gl g}^*,
%{\gl g}$, respectively, when evaluated at the identity of the group are nothing but the KSK Poisson brackets on coadjoint orbits of Lie groups.
The Lie group $D$, associated to the Lie algebra ${\gl d}= {\gl g}\bowtie {\gl g}^*$
is the Drinfel'd double group of $G$ (or $G^*$, being the construction symmetric).
%\footnote{Properly \cite{semenov} Drinfel'd doubles are the quantum version of double groups. The latter is introduced below.
%The notation classical and quantum Drinfel'd doubles is also used. }
\footnote{We denote with the symbol $\bowtie$ the Lie algebra structure of $\mathfrak{d}$ as a sum of two non-Abelian, non commuting subalgebras, each one of them acting on its dual.}
There is a dual algebraic approach to the picture above, mainly due to Drinfel'd \cite{drinfel'd}, which starts from a deformation of the semi-direct sum $\mathfrak{g}~\dot\oplus ~\R^n$, with $\R^n \simeq\mathfrak{g}^*$, into a fully non-Abelian Lie algebra, which coincides with $\mathfrak{d}$. Let us review the construction for the group $SU(2)$, whose Drinfel'd double is the group $SL(2,\C)$ \cite{drinfel'd}. }
%An action can be shown to be written on the tangent bundle of $D$, in such a way that the usual Lagrangian description of the rotator can be recovered by reducing the carrier manifold to the tangent bundle of $SU(2)$.
%\label{sl2}
%The structure of $\mathfrak{d}=\mathfrak{sl}(2,\mathbb{C})$ as a double algebra is shortly reviewed here.
%With this purpose, we start recalling that
The complex Lie algebra $\mathfrak{sl}(2)$ is completely defined by the Lie brackets of its generators:
\begin{equation}
[t_3,t_1]=2t_1; \quad [t_3,t_2]=-2t_2; \quad [t_1,t_2]=t_3;
\end{equation}
with
\begin{equation}
t_1=
\begin{pmatrix}
0 & 1 \\
0 & 0
\end{pmatrix}
; \quad t_2=
\begin{pmatrix}
0 & 0 \\
1 & 0
\end{pmatrix}
; \quad t_3=
\begin{pmatrix}
1 & 0 \\
0 & -1
\end{pmatrix}
.
\label{gensl}
\end{equation}
The real algebra $\mathfrak{sl}(2,C)$ is obtained by considering the complex linear combinations
%of the basis elements of $\mathfrak{sl}(2)$, say $e_i$, $b_i$, $i=1,2,3$, respectively given by:
\begin{equation} \label{b1}
e_1 =\frac{1}{2}(t_1+t_2)=\frac{\sigma_1}{2}, \; \;\; e_2 =\frac{i}{2}(t_2-t_1)=\frac{\sigma_2}{2}, \;\;\;
e_3 =\frac{1}{2}t_3=\frac{ \sigma_3}{2}
\ee
\be
b_i= i e_i \;\;\; i=1,2,3
\ee
We have indeed
%the real algebra $\mathfrak{sl}(2,C)$ can be easily obtained with its Lie brackets:
\be \label{su}
[e_i,e_j] = i{\epsilon_{ij}}^k e_k,~~~~
{[}e_i,b_j{]}= i{\epsilon_{ij}}^kb_k ,~~~~
{[}b_i,b_j{]}=-i{\epsilon_{ij}}^ke_k
\ee
with $\{e_i\}, i=1,2,3$, generating the $\mathfrak{su}(2)$ subalgebra.
In a similar way, one can introduce the combinations:
\begin{equation}
\tilde e^1=it_1;\qquad \tilde e^2=t_1; \qquad \tilde e^3=\frac{i}{2} t_3,
\label{basisdouble}
\end{equation}
which are dual to the generators \eqref{b1}, with respect to the scalar product naturally defined on $\mathfrak{sl}(2,\mathbb{C})$ as:
\begin{equation}
\Braket{u,v}=2\, {\rm Im}(\,Tr(uv)\,), \quad \forall u,v \in \mathfrak{sl}(2,\mathbb{C}).
\label{psd}
\end{equation}
We have indeed
\begin{equation}
\Braket{\tilde e^i, e_j}=2\, {\rm Im}(\,Tr(\tilde e^i e_j)\,)=\delta^i_j \label{eupedown}.
\end{equation}
Hence, $\{\tilde e^j\}$ span the dual vector space $\mathfrak{su}(2)^*$. This is by itself
a Lie algebra, the special Borel subalgebra $\mathfrak{sb}(2,\mathbb{C})$ with the following Lie brackets:
\begin{equation}
[\tilde e^1,\tilde e^2]=0; \qquad [\tilde e^1,\tilde e^3]=- i \tilde e^1; \qquad [\tilde e^2,\tilde e^3]=-i \tilde e^2
\end{equation}
which in compact form read
\begin{equation}
[\tilde e^i, \tilde e^j]= i {f^{ij }}_k \tilde e^k \label{sb}
\ee
with ${f^{ij}}_k=\epsilon^{ij l}\epsilon_{l3k}$.
Moreover
\begin{equation}
[\tilde e^i,e_j]= i\epsilon^i_{\,jk}\tilde e^k+ i e_k {f^{ki}}_j .
\label{liemis}
\end{equation}
For future convenience we also note that:
\be
\tilde e^i \te^j= -\frac{1}{4} \delta^{i3}\delta^{j3}\sigma_0 +\frac{i}{2} {f^{ij}}_k \te^k.\label{ee}
\ee
Since
\begin{equation}
\Braket{e_i,e_j}=\Braket{\tilde e^i,\tilde e^j}=0
\ee
both $\mathfrak{su}(2)$ and $\mathfrak{sb}(2,\mathbb{C})$ are maximal isotropic subspaces of $\mathfrak{sl} (2,\mathbb{C})$ with respect to the scalar product \eqref{psd}. We have $\mathfrak{sb}(2,\mathbb{C})= \mathfrak{su}(2)\bowtie \mathfrak{sb}(2,\C)$.
%\footnote{Notice that another splitting of the $\mathfrak{sl} (2,\mathbb{C})$ Lie algebra into maximally isotropic subspaces with respect to the same scalar product is represented by the span of $\{e_i\}, \{b_i\}, i=1,2,3$, with $\Braket{e_i, e_j}=\Braket{b_i,b_j}=0, \Braket{e_i,b_j}=\delta_{ij}$. However the generators $\{b_i\}$ do not close a Lie subalgebra.}
Therefore, the Lie algebra $\mathfrak{sl}(2,\mathbb{C})$ can be split into two maximally isotropic dual Lie subalgebras with respect to a bilinear, symmetric, non degenerate form defined on it. The couple ($\mathfrak{su}(2)$, $\mathfrak{sb}(2,\C)$), with the dual structure described above, is a Lie bialgebra. Since the role of $\mathfrak{su}(2)$ and its dual algebra can be interchanged, $(\mathfrak{sb}(2,\C)$, $\mathfrak{su}(2)$) is a Lie bialgebra as well. The triple $(\mathfrak{sl}(2,\mathbb{C}), \mathfrak{su}(2), \mathfrak{sb}(2,\mathbb{C}))$ is called a {\it Manin triple} \cite{drinfel'd}.
%The total algebra $\mathfrak{d}= \mathfrak{g}\bowtie \mathfrak{g}^*$ which is the Lie algebra defined by the Lie brackets \eqn{su}, \eqn{sb}, \eqn{liemis}, with its dual $\mathfrak{d}^*$ is also a Lie bialgebra.
%The couple $(\mathfrak{d},\mathfrak{d}^*)$ is called the {\em double} of $(\mathfrak{g},\mathfrak{g}^*)$ \cite{semenov}.
$D=SL(2,\C)$ is thus the {\em double group} obtained by exponentiating the bialgebra, endowed with some additional structures such as a Poisson structure on the group manifold compatible with the group structure. The two partner groups, $G=SU(2)$ and $G^*= SB(2,\mathbb{C})$ with suitable Poisson brackets, are named {\it dual groups}. Their role can be interchanged, so that they share the same double group $D$.
%The splitting of $\mathfrak{sl}(2,\mathbb{C})$ is realized with respect to the scalar product \eqref{psd}.
%This is given by the {Cartan-Killing metric} $g_{ij}=\frac{1}{2}{c_{ip}}^q {c_{jq}}^p$, induced by the structure constants ${c_{ij}}^k$ of $\mathfrak{sl}(2,\mathbb{C})$ in its adjoint representation.
Besides the scalar product \eqn{psd}, there is another non-degenerate, invariant scalar product, represented by
\begin{equation}
(u,v)=2 Re( \, Tr(u v) \,) \qquad \forall u,v \in \mathfrak{sl}(2,\mathbb{C}).
\label{sp2}
\end{equation}
In this case, for the basis elements, one gets:
\begin{equation}
(e_i,e_j)=\delta_{ij}, \quad (b_i,b_j)=-\delta_{ij}, \quad (e_i,b_j)=0,
\label{otherscalar}
\end{equation}
giving rise to a metric which is not positive-definite. With respect to this, new maximal isotropic subspaces can be defined, which are spanned by:
\be
f_i^+=\frac{1}{\sqrt 2} (e_i+ b_i) \,\,\, \quad ; \,\,\, \quad f_i^-= \frac{1}{\sqrt 2}(e_i-b_i) \label{newb}.
\ee
satisfying
\begin{equation}
(f^+_i,f^+_j)= (f^-_i,f^-_j)=\,\,0 \quad ; \quad (f^+_i,f^-_j)=\delta_{ij} .
\end{equation}
%whereas
%\begin{equation}
%\Braket{f^+_i,f^+_j}= \delta_{ij}, \quad \Braket{f^-_i,f^-_j}=-\delta_{ij}, \quad \Braket{f^+_i,f^-_j}=0\,\,.
%\end{equation}
By denoting by $C_+$ and $ C_-$ the two subspaces spanned by $\{e_i\}$ and $\{b_i\}$ respectively, one can notice \cite{gualtieri:tesi} that the splitting ${\mathfrak d}= C_{+} \oplus C_{-}$ defines a positive definite metric $\mathcal{H}$ on ${\mathfrak sl(2,\C)}$ via:
\be
\mathcal{H}= (\;,\;)_{C_+}- (\;,\;)_{C_-} \label{metricG}
\ee
%As in Ref. \cite{gualtieri:tesi}, the inner product is here used to identify ${\mathfrak d}$ with its dual, so that the metric $\mathcal{H}$ may be viewed as an
%automorphism of ${\mathfrak d}$ which is symmetric and which squares to the identity, i.e.
%$\mathcal{H}^2 = 1$.
Let us indicate the Riemannian metric with double round brackets. One has then:
\be
((e_i,e_j)) \equiv (e_i,e_j); ~~~~~ ((b_i,b_j)) \equiv -(b_i,b_j);~~~~~ ((e_i,b_j)) \equiv (e_i,b_j)=0 \,. \label{riem}
\ee
Let us introduce the notation:
%=\mathfrak{su}(2) \oplus \mathfrak{sb}(2,\mathbb{C})$, we can write the orthogonal elements of
%$\mathfrak{sl}(2,\mathbb{C})$ as
\begin{equation}
e_I=\begin{pmatrix}e_i\\ e^i \end{pmatrix},
\qquad e_i \in \mathfrak{su}(2), \quad e^i \in \mathfrak{sb}(2,\mathbb{C}).
\label{doubledb}
\end{equation}
Then the scalar product \eqn{psd} becomes
\begin{equation} \label{Lprod}
\Braket{e_I,e_J}={\cal \eta}_{IJ}=
\begin{pmatrix}
0 & \delta_i^j \\
\delta_j^i & 0
\end{pmatrix}
.
\end{equation}
{This symmetric inner product has signature $(d,d)$ and therefore defines the non-compact orthogonal group $O(d,d)$, with $d=3$ in this case}.
The Riemannian product \eqn{riem} yields instead:
\beqa
((\tilde e^i, \tilde e^j))&=&\delta^{ip}\delta^{jq} ((b_p + e_l{\epsilon^l}_{p3} ))((b_q+ e_k{\epsilon^k}_{j3} )) \nonumber \\
&=& \delta^{ij}+\epsilon^i_{\;l3} \delta^{lk} \epsilon^j_{\;k3} \,\,\,\,\, ; \label{sp2ee}\\
(( e_i, \tilde e^j))&=& [(e_i, b_q) + {\epsilon^k}_{q3} (e_i, e_k)] \delta^{jq}={\epsilon_{3i}}^j \,\,\, .\label{mixed}
\eeqa
Hence, one has:
\begin{equation} \label{Rprod}
((e_I,e_J))={\cal H}_{IJ}=
\begin{pmatrix}
\delta_{ij} & \delta_{il}\epsilon^{jl3 }\\
\epsilon^{il3} \delta_{lj}\; \;& \; \delta^{ij}+ \epsilon^{il3} \delta_{lk}\epsilon^{jk3}
\end{pmatrix}
.
\end{equation}
This metric satisfies the relation:
\be {\cal H}^T {\cal \eta} {\cal H}= \eta \label{compatibile}
\ee
indicating that ${\cal H}$ is a pseudo-orthogonal $O(3,3)$ matrix.
%It is interesting to see how the metric ${\cal \eta}$ in Eq. (\ref{Lprod}) and the metric ${\cal H}$ in Eq. (\ref{Rprod}) naturally emerge out in the framework here in exam. They correspond, in the usual context of DFT, to the $O(d,d)$ invariant metric and to the so-called {\em generalized metric} \cite{Tseytlin, Hohm}, respectively. In particular, in the latter, the role of the graviton field is played by the Kronecker delta $\delta_{ij} $ while the role of the Kalb-Ramond field is played by the three-dimensional Levi-Civita symbol $\epsilon_{ij3}$ with one of the indices being fixed.
\section{The Dual Model}\label{dualrot}
In this section we introduce a dynamical model on the dual group of $SU(2)$, the Borel group $SB(2,\C)$, with an action functional that is formally analogous to \eqn{lag}.
As carrier space for the dynamics in the Lagrangian (respectively Hamiltonian) formulation one can choose the tangent (respectively cotangent) bundle of the group $SB(2,\C)$.
A suitable action for the system is the following:
\be
{\tilde S}_0= \int_\R {\tilde L} _0~dt= - \frac{1}{4} \int_\R {\mathcal Tr}\,[\tilde\phi^*(\tg^{-1} d \tg\wedge *_H\tilde\phi^*( \tg^{-1} d\tg)] = - \frac{1}{4}\int_\R {\mathcal Tr} \,[(\tg^{-1}{ \dot \tg})(\tg^{-1}{ \dot \tg})] dt \label{dualag}
\ee
with $\tilde \phi : t\in \R\rightarrow SB(2,\C)$, the group-valued dynamical variable, and $\tilde\phi^*$ the pull-back map. Analogously to the IRR case,
$
\tg^{-1} d \tg= i \beta_k \te^k $ represents the Maurer-Cartan left invariant one-form on the group manifold, with $\beta_k$ the left-invariant basic one-forms, $*_H$ the Hodge
star operator on the source space $\R$, such that $*_H dt = 1$. Moreover, as previously, in order to adhere to the notation which is commonly adopted in field theory, we shall identify the dynamical variable $\tilde \phi$ with $\tg$ for the remainder of this section. The symbol ${\mathcal Tr} $ stands for a {\it suitable} scalar product on the Lie algebra $\mathfrak{sb}(2,\C)$. Indeed, since the algebra is not semi-simple, there is no scalar product which is both non-degenerate and invariant. We choose to work with the scalar product induced by the Riemannian metric $\mathcal{H}$, which, on the algebra $\mathfrak{sb}(2,\C)$ takes the form \eqn{sp2ee} which is positive definite and non-degenerate. This is only invariant under left $SB(2,\C)$ action besides being $SU(2)$ invariant.
Indeed, by observing that the generators $\te^i$ are not Hermitian, \eqn{sp2ee} can be verified to be equivalent to:
\vspace{0.3cm}
\be
((u,v)) \equiv 2{\rm Re}\Tr [(u)^\dag v] \label{2ndprod}
\ee
which
is not invariant under right $SB(2,\C)$ action.
Similarly to the IRR case, the model can be regarded as a $(0+1)$-dimensional field theory which is
group-valued.
The group manifold can be parametrized with $\R^4$ coordinates as $\tg= 2( u_0 \te^0
+ i u_i \te^i)$, with $u_0^2- u_3^2=1$ and $\te^0= {\mathbb I}/2$.
%One has then:
%\be
%u_i=\frac{1}{4} (( i\tg, \te^i)), \;i=1,2,~~~~ \; u_3=\frac{1}{2} (( i\tg, \te^3)), ~~~~\; u_0= \frac{1}{2}((\tg, \te^0)) \nonumber \;\;\;
%\ee
%where the last product is defined as twice the real part of the trace, in order to be consistent with the others.
By observing that
\be
\tg^{-1} \dot \tg=2 i (u_0\dot u_i-u_i \dot u_0+ {f_{i}}^{\,jk} {u_j \dot u_k })\te^i \label{tiqdot}
\ee
the Lagrangian in \eqn{dualag} can be rewritten as:
\be
\tilde{{L}}_0= (u_0\dot u_i-u_i \dot u_0+ {f_{i}}^{\,jk} {u_j \dot u_k })(u_0\dot u_r-u_r \dot u_0+ {f_{r}}^{\,pq} {u_p \dot u_q })((\te^i,\te^r))
= \dot \tQ_i \dot \tQ_r h^{ir} \nonumber
\ee
with
$
\dot \tQ_i \equiv u_0\dot u_i-u_i \dot u_0+ {f_{i}}^{\,jk} {u_j \dot u_k } \nonumber
$
the left generalized velocities
and
\be
h^{ir} \equiv (\delta^{i r}+ {\epsilon^i}_{l3}{\epsilon^r}_{s3}\delta^{ls}) \label{hir}
\ee
the metric defined by the scalar product. $(\tQ_i, \dot \tQ_i)$ are therefore tangent bundle coordinates, with $\tQ_i$ implicitly defined.
By repeating the analysis already performed for the IRR, one finds the equations of motion:
\be
{\sf L}_\Gamma(\dot \tQ_j\, i_{\tX^i}\beta_l )h^{jl} - {\sf L}_{\tX^i} \tilde{L}_0 = {\sf L}_\Gamma\dot \tQ_j h^{ji} - \dot \tQ_p \dot \tQ_q {f^{ip}}_k h^{qk} = 0. \label{LGamma}
\ee
with $\tX^j$ being the left invariant vector fields generating the right action of $SB(2,\C)$. Differently from the IRR case, the second term in the RHS is not vanishing, because the structure constants are not completely antisymmetric. This is to be expected because the Lagrangian is not invariant under right action.
It has to be noticed here that, analogously to the IRR case, one could define the right generalized velocities on the fibers starting from right invariant one-forms, but, differently from that case, the right invariant Lagrangian is not equivalent to the left invariant one.
The cotangent bundle coordinates are $(\tQ_i, \tI^i)$ with $\tI^i$ the conjugate left momenta
$
\tI^j= \frac{\del {\tilde{{ L}}}_0}{\del \dot \tQ_j}= \dot \tQ_r (\delta^{j r}+ \epsilon^j_{\,l3}\epsilon^r_{s3}\delta^{ls}).
% = \frac{i}{2} ((\tg^{-1}\dot\tg,\te^i))\delta_{i}^j \, .
$
On inverting for the velocities $\dot\tQ_j= \tI^i(\delta_{ij}-\frac{1}{2}\epsilon_i^{\,p3}\epsilon_j^{\,q3}\delta^{pq}), $ we get the Hamiltonian function:
\be
\tilde{H}_0=[\tI^j \dot \tQ_j -\tilde L]_{\dot \tQ=\dot \tQ(\tI)}= \frac{1}{2}\tI^i (h^{-1})_{ij }\tI^j \,\,\, ,\label{h0du}
\ee
with
\be
(h^{-1})_{ij } \equiv (\delta_{ij}-\frac{1}{2}\epsilon_i^{\,p3}\epsilon_j^{\,q3}\delta_{pq}).
\label{hinvij}
\ee
%the inverse of Eq. \eqn{hir}.
On introducing the linear combination:
$
\tilde I= i \tilde I^j {\te_j}^*
%\label{tIform}
$
with $\langle {e_j}^* |\te^i\rangle=\delta_j^i$ we obtain
the first order dynamics by means of the following Poisson brackets:
\beqa
\{u_i,u_j\}&=&0\label{ppdu}\\
\{\tI^i,\tI^j\}&=& {f^{ij\;}}_k\tI^k \label{xxdu}\\
\{u_i,\tI^j\}&=&{\delta_{i}^j u_0 -{f_i}^{\; jk}u_k \;\;\; {\rm or ~ equivalently}\;\;\; \{\tg, \tI^j\}= 2 i \tg \te^j }\label{ijdu}
\eeqa
which are derived from the first order formulation of the action functional (see \cite{MPV18} for details).
%Since the results are slightly different from the IRR case, let us present the derivation in some detail.
%The first-order action functional reads in this case as:
% \be
%\tilde S_1= \int \langle \tI| \tg^{-1}d \tg\rangle - \int \tilde H dt \equiv \int \tilde \vartheta -\int \tilde H dt \, .
%\ee
%Observing that:
%\be
%\langle \tI| \tg^{-1}d \tg\rangle =i \tI^i \delta^k_i \beta_k
%\ee
%the symplectic form $\tilde \omega$ reads as:
%\be
%\tilde \omega= d \tilde \vartheta= d \tI^j \wedge \beta_j - \frac{1}{2}\tI^j {f_j}^{lm} \beta_l\wedge \beta_m \label{tomega}
%\ee
%where the relation
% $d [\tg^{-1}d \tg] = \frac{i}{2} \beta_i\wedge\beta_j { f^{ij}}_k\te^k$ has been used. By inverting $\tilde \omega$, one finally finds the Poisson algebra \eqn{ppdu}-\eqn{ijdu}.
Specifically we get:
\be
\dot\tI^j= \{\tI^j,\tilde H\}= f^{jk}_l\tI^l\tI^r h^{-1}_{kr}
\ee
which is different from zero, as expected, expressing the non-invariance of the Hamiltonian under right action.
Vice-versa, by introducing the right momenta $\tJ^i$
one readily obtains:
\be
\dot\tJ^j= \{\tJ^j,\tilde H\}= 0
\ee
namely, right momenta are constants of the motion and the Hamiltonian is invariant under left action, as it should.
%By using \eqn{ijdu} it is possible to find:
%\be
%\tg^{-1}\dot \tg= 2 i \te^i (h^{-1})_{ij} \tI^j \nonumber
%\ee
%consistently with Eq. \eqn{tiqdot}.
Right momenta are therefore conserved, as for the rigid rotator, while left momenta are not.
The same remark as at the end of Sect. \ref{rigidrot} applies: while the fibers of the tangent bundle $TSB(2,\C)$ can be identified, as a vector space, with the Lie algebra of $SB(2,C)$, $\mathfrak{sb}(2,\C)\simeq \R^3$, with $\dot \tQ_i$ denoting vector fields components, the fibers of the cotangent bundle $T^*SB(2,C)$ are isomorphic to the dual Lie algebra $\mathfrak{sb}(2,\C)^*$. As a vector space this is again $\R^3$, but $\tI^j$ are now components of one-forms.
As a group, $T^*SB(2,\C)$ is the semi-direct product of $SB(2,\C)$ and the Abelian group $\mathbb{R}^3$, with Lie algebra the semi-direct sum represented by
\be
\left[B_i,B_j\right] = i {f_{ij}}^k B_k , ~~~~~
\left[S_i,S_j\right] = 0, ~~~~~
\left[B_i,S_j\right] = i {f_{ij}}^k S_k. \label{BS}
\ee
Then, as for the IRR, the non-triviality of the Poisson algebra over the cotangent bundle of the group $SB(2,\C)$ reflects the structure of the latter.
% on $\mathfrak{sb}(2,\C)^*\simeq \mathbb{R}^3$ , i.e. its dual algebra, and it reflects the non-triviality of the Lie bracket \eqn{BB} with the Lie algebra generators $B_i$ identified with linear functions on the dual algebra.
Before closing the section, let us summarize the results. We have introduced a model on the dual group of $SU(2)$, the Borel group $SB(2,\C)$, whose Hamiltonian dynamics is retrieved in terms of Poisson brackets of KSK type. As we shall see, the Poisson brackets of the momenta $I_i, \tI^i$ are dually related.
\section{The generalized action}\label{gensec}
In the previous sections we have introduced two dynamical models on configuration spaces which are dual Lie groups. The Poisson algebras for the respective cotangent bundles, $T^*SU(2)$, $T^*SB(2,\C)$, which we restate for convenience in the form:
\beqa
\{g,g\}&=&0,\;\;\;\;\{I_i,I_j\}= {\epsilon_{ij}}^k I_k , \;\;\;\;\; \{g, I_j\}= 2 i g e_j \label{sudue}\\
\{\tg,\tg\}&=&0,\;\;\;\;
\{\tI^i,\tI^j\}= {f^{ij}}_k\tI^k ,\;\;\;\;
\{\tg, \tI^j\}= 2 i \tg \te^j \,\,\, , \label{sbdue}
\eeqa
have both the structure of a semi-direct sum which reflects the semi-direct structure of the Lie algebras $\mathfrak{su}(2)\dot\oplus\R^3$ and $\mathfrak{sb}(2,\C) \dot \oplus \R^3$.
%By identifying the dual algebras $\R^3$, in both cases, with an Abelian Lie algebra, we have that each semi-direct sum has the the form \eqn{Liesuma}, with $\R^3$ generators satisfying trivial brackets and with a trivial $ad^*$ action:
%\be
%[X+\xi,Y+\zeta]=[X,Y]-ad^{*}_{X}\zeta + ad^{*}_{Y}\xi \label{sumatriv} \,\,\,.
%$\ee
%To this, it is sufficient to expand the group variables, $g,\tg$
%$$ g\simeq \mathds{1}+i \lambda J^i e_i + O(\lambda^2), \;\;\;\; \tg\simeq \mathds{1}+i \mu \tilde{J_i} \te^i + O(\mu^2)$$ \label{groupvariables}
%and compute the related Poisson brackets in Eqs. \eqn{sudue}, \eqn{sbdue}
%to first order in the parameters. One gets:
%\beqa
%\{J^i, J^j\}&=& 0,\;\;\;\;\{I_i,I_j\}= {\epsilon_{ij}}^k I_k , \;\;\;\;\;\{J^i, I_j\}= -{\epsilon^{i}}_{jk} J^k \label{suduee}\\
%\{\tilde {J}_i, \tilde {J}_j\}&=& 0,\;\;\;\;\{\tI^i,\tI^j\}= {f^{ij}}_k\tI^k ,\;\;\;\;\{\tilde {J}_i, \tI^j\}= - \tilde{J}^k{f_{ki}}^j.\label{sbduee}
%\eeqa
%The new dynamical variables $J^i, \tilde {J}_i$ will be identified in the forthcoming section with $I^i, \tI_i$ by unifying the cotangent bundles $T^*SU(2), T^*SB(2,\C)$ into the Drinfel'd double $SL(2,\C)$. The brackets \eqn{suduee}, \eqn{sbduee} will then emerge naturally as appropriate limits of the Poisson-Lie brackets on the dual groups $G$, $G^*$, when evaluated at the identity of the respective groups as in Eq. \eqn{Liesuma}.
In order to unify the two models within a generalized action, whose configuration space has {\it double} dimension with respect to the previous ones, let us introduce the configuration space variable $\Phi:t\in \mathbb{R} \rightarrow \gamma \in SL(2,\mathbb{C})$. The left invariant one-form on the group manifold is then:
\begin{equation}
\Phi^*(\gamma^{-1} \mathrm{d}\gamma)= \gamma^{-1} \dot \gamma\; dt \equiv \dot {\bf Q}^I e_I \mathrm{d}t \label{gammagamma}
\ee
with $e_I=(e_i, \tilde e^i)$ the $\mathfrak{sl}(2,\C)$ basis introduced in Eq. \eqn{doubledb} and $ \dot {\bf Q}^I$, the left generalized velocities. As in the previous sections, in order to adhere to the notation which is commonly adopted in field theory, we shall identify the dynamical variable $\Phi$ with $\gamma$ for the remainder of this section. By defining the decomposition $\dot {\bf Q}^I \equiv( A^i, B_i)$ one has:
\be
\gamma^{-1} \dot \gamma \;dt= (A^i e_i + B_i \tilde e^i) dt \nonumber
\ee
where, however, both components are tangent bundle coordinates for $SL(2,C)$.
%\footnote{We could alternatively interpret $(A^i, B_i)$ as fiber coordinates of the generalized bundle $T\oplus T^*$, with base manifold $SU(2)$, so that the model is both an instance of Generalized Geometry or of Doubled Geometry.}.
By using the scalar product \eqn{psd}, the components of the generalized velocity can be explicitly obtained:
\be
A^i= 2{\rm Im} \Tr ( \gamma^{-1}\dot \gamma \tilde e^i); \;\;\; B_i= 2{\rm Im} \Tr (\gamma^{-1}\dot \gamma e_i). \nonumber
\ee
%Since $*\gamma^{-1} \mathrm{d}\gamma= \dot {\bf Q}^I e_I $ with , $* dt = 1$,
The proposed action \cite{MPV18} is the following:
\begin{equation}
{S}= \int_R {L} dt= \frac{1}{2}\int_{\mathbb{R}}\bigl( k_1\Braket{\gamma^{-1}\mathrm{d}\gamma\stackrel{\wedge}{,}* \gamma^{-1}\mathrm{d}\gamma} +k_2 ((\gamma^{-1}\mathrm{d}\gamma \stackrel{\wedge}{,} * \gamma^{-1}\mathrm{d}
\gamma)) \bigr),
\label{newac}
\end{equation}
where $k_1,k_2$ are real parameters, and the two non-degenerate scalar products in Eqs. \eqn{Lprod}, \eqn{Rprod}, are employed, yielding
\beqa
\Braket{\gamma^{-1}\mathrm{d}\gamma \stackrel{\wedge}{,} * \gamma^{-1}\mathrm{d}\gamma}
&=& \dot {\bf Q}^I \dot {\bf Q}^J \eta_{IJ}\\
((\gamma^{-1}\mathrm{d}\gamma\stackrel{\wedge}{,} *\gamma^{-1}\mathrm{d}\gamma))&=& \dot {\bf Q}^I \dot {\bf Q}^J \mathcal{H}_{IJ}
\label{prodotto2}
\eeqa
so that the generalized Lagrangian reads
\be \label{explito}
{L}= \frac{1}{2} ( k\, {\cal \eta}_{IJ} + {\cal H}_{IJ})\dot {\bf Q}^I \dot {\bf Q}^J
\ee
with
\be
k\, {\cal \eta}_{IJ}+ {\cal H}_{IJ}=
\begin{pmatrix}
\delta_{ij}& k \delta_i^j + {\epsilon_{3i}}^{j} \\
-{\epsilon^i}_{j3}+k \delta_i^j& \delta^{ij}+ {\epsilon^i}_{l3} \epsilon^j_{k3}\delta^{lk}
\end{pmatrix} \nonumber
\ee
and where the position $k_1/k_2 \equiv k $ has been made.
%This leads to:
%\be \label{newlag}
%{L} = \frac{1}{2} \bigl[ \delta_{ij} A^i A^j + ( k \delta_{i}^j+ {\epsilon_{i}}^{j3}) A^i B_j + ( k \delta^{i}_j- {\epsilon^{i}}_{j3}) B_i A^j + ( \delta^{ij}+\delta^{lk} \epsilon^i_{l3} \epsilon^j_{k3}) B_i B_j \bigr] .
% \ee
%The Lagrangian one-form is therefore:
%\be
%{\boldsymbol{\theta}}_{L}=(k\, {\cal \eta}_{IJ}+ {\cal H}_{IJ})\dot {\bf Q}^I {\boldsymbol{\alpha}}^J
%\ee
Proceeding as before we get the equations of motion in the form:
\be
{\sf L}_\Gamma\dot {\bf Q}^I( k\, {\cal \eta}_{IJ}+ \, {\cal H}_{IJ}) -\dot {\bf Q}^P \dot {\bf Q}^Q C_{IP}^K ( k\, {\cal \eta}_{QK}+ \,{\cal H}_{QK}) =0 \label{eomd}
\ee
where $C_{IP}^K$ are the structure constants of $\mathfrak{sl}(2,\C)$. The matrix $ k\, {\cal \eta}_{IJ}+ \, {\cal H}_{IJ}$ is non-singular, provided $k^2 \ne 1$, which will be assumed from now on.
In \cite{MPV18} it has been shown that the Lagrangian of the IRR and of its dual model can be recovered by exploring the global symmetries of the generalized dynamics. If we choose a local parametrization for the elements of the double group $SL(2,\mathbb{C})$: $\gamma= \tilde{g}g$, with $g \in SU(2)$ and $\tilde{g} \in SB(2,\mathbb{C})$,
from Eq. \eqref{newac} it is easily seen that $ L$ is invariant under left and right action of the group $SU(2)$,
and under left action of the group $SB(2,\mathbb{C})$. In order to recover the IRR Lagrangian we therefore gauge the $SB(2,\C)_L$ global symmetry, by introducing a $SB(2,\C)$ gauge connection $\tilde C= \tilde {C_i}(t)\tilde e^i $:
\be\label{sbgauge}
\gamma^{-1}d\gamma\rightarrow \gamma^{-1}D_{\tilde C}\gamma= (\gamma^{-1}\dot\gamma+\gamma^{-1}\tilde C\gamma) dt
\ee
and performing the substitution
\be
\gamma^{-1}\dot\gamma+\gamma^{-1}\tilde C\gamma=\gamma^{-1}\dot\gamma+ \tilde{C_i} \gamma^{-1} \tilde e^i \gamma= \mathcal{U}_i\tilde e^i + \mathcal{W}^i e_i \nonumber \,\,\,
\ee
from which the new variables $\mathcal{U}_i, \mathcal{W}^i$ are easily retrieved.
Analogously, in order to obtain the Lagrangian of the dual model we gauge the global $SU(2)_R$ invariance, by introducing the $SU(2)$ gauge connection $C= C^i (t) e_i$ so to have
\be
\gamma^{-1}d\gamma\rightarrow \gamma^{-1}D\gamma=\tilde {\mathcal U}_i {\tilde e}^i + \tilde {\mathcal W}^i e_i .
\ee
By considering each of the gauged Lagrangian functions, $L_{\tilde C}$ or $L_{C}$ and re-expressing them in terms of the new variables,
the partition function $Z$ is considered for each of them
\begin{equation}
Z_1=\int \mathcal{D}g \mathcal{D}\tilde{g}\mathcal{D} {\tilde C} e^{-S_{{\tilde C}}}
\label{partitt}
\end{equation}
or
\begin{equation}
Z_2=\int \mathcal{D}g \mathcal{D}\tilde{g}\mathcal{D} { C} e^{-S_{{C}}}
\label{partitt}
\end{equation}
and the integration over the gauge potential $\tilde C$, respectively $C$, is performed. Using techniques which are standard in field theory, the integration with respect to the gauge potentials is traded for the integration with respect to ${\mathcal U}_i$, $\tilde {\mathcal{W}}^i $ respectively, so that we are left with half the degrees of freedom of the generalized action \eqn{newac} and we retrieve the IRR model or the dual model, depending on which gauged Lagrangian we started with. We refer for details to \cite{MPV18}.
A generalized kinematics in the context of DFT has been considered in \cite{freidel}.
\subsection{The Hamiltonian Formalism}\label{hamform}
In the doubled description introduced above, the left generalized momenta are represented by:
\be
{\bf P}_I = \frac{\del L}{\del \dot {\bf Q}^I}= ( {\cal \eta}_{IJ}+ k\,{\cal H}_{IJ})\dot {\bf Q}^J \label{genP}
\ee
The Hamiltonian reads then as:
\be
{H}= ({\bf P}_I \dot{\bf Q}^I - L)_{{\bf P}}= \frac{1}{2} [( {\cal \eta}+ k\, {\cal H})^{-1}]^{IJ} {\bf P}_I{\bf P}_J \nonumber
\ee
with
\be
[( {\cal \eta} + k\, {\cal H})^{-1}]^{IJ}= \frac{1}{2} (1-k^2)^{-1}
\begin{pmatrix}
\delta^{ij}+ \epsilon^i_{l3} \epsilon^j_{k3}\delta^{lk}& -{\epsilon^i}_{j3}-k \delta^i_j
\\
{\epsilon_i}^{j3}-k \delta_i^j& \delta_{ij} \,\,\,
\end{pmatrix} \,\,. \nonumber
\ee
From \eqn{genP} one can explicitly write the generalized momenta ${\bf P}_I$ in terms of the components of $\dot{\bf Q}^I\equiv(A^i, B_j)$, finding:
\be
{\bf P}_I \equiv ( I_i, \tI^i)=\left(\delta_{ij} A^j+(k\delta_i^j+ \epsilon_i^{j3})B_j, (k \delta^i_j-\epsilon^i_{j3})A^j+[\delta^{ij}+\delta^{lk}\epsilon^i_{l3}\epsilon^j_{k3}]B_j\right). \nonumber
\ee
In terms of the components $I_i, \tI^j$, it turns out that:
\be
{H}
%&=&\frac{1}{2}(1-k^2)^{-1}\left( \delta^{ij} I_i I_j + \delta_{ij} \tI^i \tI^j + \epsilon^i_{l3} \epsilon^j_{k3}\delta^{lk} I_i I_j -2 k \delta^i_j I_i \tI^j + 2\epsilon_i^{j3} \tI^iI_j \right) \nonumber \\
=
\frac{1}{2}(1-k^2)^{-1} \left ( (1-k^2) \delta^{ij} I_i I_j + \delta_{ij}(\tI^i -I_s(k \delta^{si}+ {\epsilon^{si}}_3))
(\tI^j -I_r(k \delta^{rj}+ {\epsilon^{rj}}_3))\right)\nn
\ee
%which can be rewritten as
%\be
%{{H}} =\frac{1}{2}(1-k ^2)^{-1}\left((1-k ^2) \delta^{ij} I_i I_j + \delta_{ij}\tilde{\cal I}^i \tilde{\cal I}^j\right) \nonumber
%\ee
%after having defined
%\be
%\tilde{\cal I}^i \equiv \tI^i -I_s(k \delta^{si}+ {\epsilon^{si}}_3)= \delta^{ij} (1-k^2) B_j. \nonumber
%\ee
In order to obtain the Hamilton equations for the generalized model on the Drinfel'd double, one can proceed as in the previous sections with the determination of Poisson brackets from the first-order action functional \cite{MPV18}, which yields:
%\be
%{{\mathcal{S}}}= \int \langle {\bf P} | \gamma^{-1}d \gamma\rangle - \int \widehat{H} dt \equiv \int {\boldsymbol{\theta}} -\int \widehat{H} dt \nonumber
%\ee
%with
%\beqa
%{\bf P}&=& i\; {\bf P}_I {e^I}^*= i\;(I_i{e^i}^* + \tI_i \te_i^*) \nonumber \\
%\gamma^{-1}d\gamma&=& i\, {\boldsymbol{\alpha}}^J e_J= (\alpha^k e_k + \beta_k \te^k) \nonumber \,\, .
%\eeqa
%We stress once again that ${\bf P_I}$, $ {\boldsymbol{\alpha}}^J$ are respectively generalized momenta and basis one-forms on the doubled configuration space $SL(2,\C)$. The symplectic form on $T^*SL(2,\C)\simeq SL(2,\C)\times \mathfrak{sl}(2,\C)^*$ is therefore:
%\beqa
%{\boldsymbol{\omega}}= d {\boldsymbol{\theta}}&=& dI_i\wedge \alpha^i + d\tI^i\wedge \beta_i +\frac{1}{2}\tI^l\left(\alpha^j\wedge\beta_k {\epsilon^k}_{jl}- \beta_j\wedge \alpha^k {\epsilon^j}_{kl}- \beta_j\wedge\beta_k {f^{jk}}_l\right)\nn\\
%&+& \frac{1}{2}I_l\left(-\alpha^j\wedge\alpha^k {\epsilon^l}_{jk}+ \alpha^j\wedge \beta_k {f^{lk}}_{j}- \beta_j\wedge\alpha^k {f^{lj}}_k\right) \nonumber
%\eeqa
%which yields for the generalized momenta the Poisson brackets:
\beqa \label{remark}
\{I_i, I_j\}&=& {\epsilon_{ij}}^k I_k \\
\{\tI^i, \tI^j\}&=& {f^{ij}}_k \tI^k\\
\{I_i, \tI^j\}&=& {\epsilon^j}_{il} \tI^l- I_l {f^{lj}}_i \;\;\;\;\{\tI^i, I_j\}= -{\epsilon^i}_{jl} \tI^l+ I_l {f^{li}}_j \label{remark3}
\eeqa
while the Poisson brackets between momenta and configuration space variables $g,\tg$ are unchanged with respect to $T^*SU(2), T^*SB(2,\C)$. We refer to \cite{MPV18} for details. %We shall come back to the Poisson algebra \eqn{remark} in the next subsection.
Poisson brackets may be written in compact form:
%In order to derive Hamilton equations, it is sufficient to write in compact form:
\be
\{{\bf P}_I, {\bf P}_J\}= C_{IJ}^K {\bf P}_K \nonumber
\ee
with $C_{IJ}^K$ the structure constants specified above. Thus we have, for Hamilton equations of motion:
\be
\frac{d}{dt} {\bf P}_I= \{ {\bf P}_I, \widehat H\}= [( {\cal \eta}+ k\, {\cal H})^{-1}]^{JK} \{ {\bf P}_I, {\bf P}_J\} {\bf P}_K= [( {\cal \eta}+ k\, {\cal H})^{-1}]^{JK} C_{IJ}^L {\bf P}_L{\bf P}_K \nonumber
\ee
which is not zero, consistently with \eqn{eomd}.
\subsection{The Poisson Algebra}\label{canform}
The generalized formulation of the isotropic rotator is completed by discussing the Poisson brackets on the double group $SL(2,\C)$, which correctly generalize those on the cotangent bundle stated in Eqs. \eqn{pp}-\eqn{ij} as well as in Eqs. \eqn{ppdu}-\eqn{ijdu}. These have been introduced long time ago in \cite{semenov, alex} in the form
\be
\{\gamma_1,\gamma_2\}= -\gamma_1\gamma_2 r^* -r \gamma_1\gamma_2 \label{gammagamma2}
\ee
where $\gamma_1= \gamma\otimes 1, \gamma_2= 1\otimes \gamma_2$ while $r \in \mathfrak{d} \otimes \mathfrak {d}$ is the classical Yang-Baxter matrix:
\be
r = e^i\otimes e_i \label{rmatrix}
\ee
satisfying the modified Yang-Baxter equation
\be
[r_{12},r_{13}+r_{23}]+ [r_{13},r_{23}]= h \nonumber
\ee
with $r_{12 }=e^i\otimes e_i \otimes \mathds{1}$, $r_{13}=e^i\otimes\mathds{1}\otimes e_i $, $r_{23}= \mathds{1\otimes e^i\otimes e_i }$, and $h\in \mathfrak{d}\otimes \mathfrak{d}\otimes\mathfrak{d}$ and adjoint invariant element in the enveloping algebra. The matrix
\be
r^*= - e_i\otimes e^i \label{rmatrix2}
\ee
is also solution of the Yang-Baxter equation.
%The group $D$ equipped with the Poisson bracket \eqn{gammagamma2} is also called the Heisenberg double \cite{semenov,alex}.
On writing $\gamma$ as $\gamma= \tilde g g$ it can be shown that \eqn{gammagamma2} is compatible with the following choice
\begin{align}
\{g_1,g_2\} &= [r^*,g_1g_2],
\label{pbm1}\\
\{{\tilde g}_1,g_2\} &=- {\tilde g}_1r g_2 \label{finalpoi}\\
\{\tilde g_1,\tilde g_2\} &=-[r,\tilde g_1\tilde g_2],
\label{pbm2}
\end{align}
with $g_1=g\otimes \mathds{1}$, $g_2=\mathds{1}\otimes g$, $\tilde g_1={\tilde g}\otimes \mathds{1}$ and ${\tilde g}_2= \mathds{1} \otimes {\tilde g}$. Eqs. \eqn{pbm2} \eqn{pbm1} are the so-called Sklyanin brackets \cite{skly}. We also have $\{ { g}_1,{\tilde g}_2\} =- {\tilde g}_2 r^* g_1$, with \eqn{pbm1}, \eqn{pbm2} Poisson-Lie brackets on $G$, $G^*$ respectively. Let us sketch how to recover Eqs. \eqn{pp}-\eqn{ij} when passing from the double group $SL(2,\C)$ to either of the cotangent bundles. In order to recover the Poisson algebra for $T^*SU(2)$, one has to rescale the matrix $r$ and the group elements of $SB(2,\C)$ by a real parameter $\lambda$. By expanding up to first order, $
\tilde g(\lambda)=e^{i\lambda I_i e^i} = 1+i\lambda I_i e^i + \mathcal{O}(\lambda^2)
$ and replacing into \eqref{pbm2} we obtain:
\begin{equation}
\{\tilde g_1,\tilde g_2\}=\simeq -\lambda^2 e^i\otimes e^j \{I_i,I_j\} + \mathcal{O}(\lambda^3), \nonumber
\end{equation}
which has to be equated to
\be
[r,\tilde g_1 \tilde g_2]\simeq - \lambda^2 I_k \epsilon^k_{\,ij}e_i\otimes e_j+ \mathcal{O}(\lambda^3)
\ee
thus yielding
\begin{equation}
\{I_i,I_j\}=\epsilon^k_{\,ij}I_k. \label{first}
\end{equation}
As for Eq. \eqn{finalpoi} in order to compute its l.h.s. we use the parametrization $g= y^0 \sigma_0 + i y^i \sigma_i$, so that, up to first order in $\lambda$
\be
\{\tilde g_1,g_2\}=2 i\lambda\left( \{I_i, y^0\} e^i\otimes e_0+i \{I_i, y^j\} e^i\otimes e_j\right)+ O(\lambda^2)\label{lhs2}
\ee
while for the r.h.s.
\be
-\tilde g_1 r g_2\simeq - \lambda( y^0 e^k\otimes e_k+ i y^j e^k\otimes (\delta_{kj}e_0+ i \epsilon_{kj}^i e_i)\label{rhs2}
\ee
thus yielding
\beqa\label{second}
\{I_i, y^0\}&=& - y^j \delta_{ij} \nonumber \\
\{I_i, y^j\}&=& y^0 \delta_i^j - y^k \epsilon_{ki}^j \nonumber
\eeqa
where the first one is compatible with the second one, by using $(y^0)^2= 1- \sum_k y^k y^k$.
Finally, on considering \eqref{pbm1} we observe that the LHS doesn't depend on $\lambda$ whereas the RHS does. Therefore we get
\be
\{y^0, y^j\} = \{y^i, y^j\} = 0 + O(\lambda) \,\,\, . \label{third}
\ee
Thus, Eqs. \eqn{first}, \eqn{second}, \eqn{third} reproduce correctly the canonical Poisson brackets on the cotangent bundle i$T^*SU(2)$.
The Poisson brackets for the cotangent bundle $T^*SB(2,\C)$ are obtained in complete analogy, when considering $r^*$ as an independent solution of the Yang-Baxter equation
\be
\rho= -\mu e_k\otimes e^k \label{altrmat}
\ee
and expanding $g\in SU(2)$ as a function of the parameter $\mu$.
\be
g= \mathds{1} + i \mu \tI^i e_i + O(\mu^2) \label{gexp} \,\,\, .
\ee
%By repeating the same analysis as above, it is straightforward to prove that the Poisson structure induced by $\rho$ is the one that correctly reproduces the canonical Poisson brackets on the cotangent bundle of $G^*=SB(2,C)$ derived in Eqs. \eqn{ppdu}-\eqn{ijdu}. Indeed, by substituting \eqn{gexp} in the LHS of \eqn{pbm1} one finds:
%\begin{equation}
%\{ g_1, g_2\}\simeq -\mu^2 e_i\otimes e_j \{\tI^i,\tI^j\} + \mathcal{O}(\mu^3), \nonumber
%\end{equation}
%and for the right-hand side:
%\begin{equation}\label{calcdu}
%\begin{split}
%[\rho, g_1 g_2]\simeq & -\mu \left([e_i, i \mu \tI^j e_j] \otimes e^i + e_i \otimes [e^i, i \mu \tI^j e_j]\right) + \mathcal{O}(\mu^3) \\
%= & -i \mu^2 \tI^j \bigl([e_i,e_j]\otimes e^i + e_i \otimes [e^i,e_j]\bigr)+ \mathcal{O}(\mu^3) \\
%= & \mu^2 \tI^j (\epsilon_{ij}^r e_r \otimes e^i + { f^{ri}}_j e_i\otimes e_r+{\epsilon^i}_{jr} e_i\otimes e^r)+ \mathcal{O}(\mu^3) \\
%= & \mu^2 \tI^k {f^{\,ij}}_k e_i\otimes e_j+ \mathcal{O}(\mu^3).
%\end{split}
%\end{equation}
%By equating the two sides, in the limit $\mu \rightarrow 0$, one obtains the Poisson bracket:
%\begin{equation}
%\{\tI^i,\tI^j\}=f_k^{\,ij}\tI^k. \label{scnd}
%\end{equation}
\subsection{Poisson-Lie simmetries}\label{PLsym}
Let us explicitly address the nature of symmetries of the dual models introduced in the previous sections. In particular we want to discuss to what extent the models possess Poisson-Lie symmetries. We closely follow \cite{marmo:articolo1} for this subsection. Poisson-Lie symmetries are Lie group transformations implemented on the carrier space of the dynamics via group multiplication, which, in general, are not canonical transformations as they need not preserve
the symplectic structure. However, if the Poisson structure is of the form \eqn{gammagamma2} with carrier space $D$ itself, or \eqn{pbm1}, \eqn{pbm2} if we are looking at $G$, $G^*$ respectively, Poisson brackets can be made invariant if the parameters of the group of transformations are imposed to have
nonzero Poisson brackets with themselves. Group multiplication is then said to
correspond to a Poisson map. We have for example, for the
right transformations of $G$ on
$D$,
\be\label{rightGact}
\gamma\rightarrow \gamma h \;,\; h \in G\;,\; \gamma \in D
\ee
and the left action of $G^*$ on $D$,
\be\label{leftG*act}
\gamma\rightarrow \tilde h \gamma \;,\; h^* \in G^*\; \; \gamma \in D
.
\ee
In terms of the coordinates $(\tilde g,g)$ this implies
\be
g\rightarrow gh\;, \quad \tilde g\rightarrow \tilde g \;,
\ee
for the former and
\be
g\rightarrow g\;, \quad \tilde g\rightarrow \tilde h \tilde g \;,
\ee
for the latter. By themselves these
transformations do not preserve the Poisson brackets
\eqn{pbm1}-\eqn{pbm2}. But they can be made to be invariant
if we require that the parameters of the transformation, $h$, have the following Poisson brackets
\be\label{hh}
\{h_1,h_2\}=[\;r^*\;,\;h_1 h_2\;] \;,
\ee
and zero Poisson brackets with $g$ and $\tilde g$.
Then $SU(2)$ right multiplication is a Poisson map and \eqn{rightGact} corresponds
to a Poisson Lie group transformation.
For \eqn{leftG*act} to be a Poisson Lie group transformation, $\tilde h$
must have the following Poisson bracket with itself
\be\label{hsthst}
\{\tilde h_1,\tilde h_2\}=-[\;r\;,\;\tilde h_1 \tilde h_2\;] \;,
\ee
and zero Poisson brackets with $g$ and $\tilde g$.
Since the right-hand-sides of \eqn{hh} and \eqn{hsthst} vanish
in the limit $\lambda \rightarrow 0 $,
the transformations \eqn{rightGact} and \eqn{leftG*act} become canonical in the limit.
\\
Moreover, Poisson brackets \eqn{pbm1}-\eqn{pbm2} are invariant under the
simultaneous action of both $G$ and $G^*$ via \eqn{rightGact} and \eqn{leftG*act}, if
we assume that{}\be
\{\tilde h_1,h_2\}=0\;.
\ee
By comparing with eq. \eqn{finalpoi} we
conclude that the algebra of the observables
$g$ and $\tilde g$ is different from the algebra of the symmetries
parametrized by $h$ and $\tilde h$.
Therefore, dynamics on the group manifold of $SL(2,\C)$ and on the two partner groups $SU(2)$ and $SB(2,\C)$ possesses Poisson-Lie group symmetries, when endowed with the above mentioned brackets.
Let us go back to the symplectic structures of the IRR and the dual model, respectively given by Eqs. \eqn{xx} and \eqn{xxdu}. The former is obtained from \eqn{pbm2} while the latter is obtained from \eqn{pbm1}, for small (but non-zero) value of the parameters $\lambda$ and $\mu$, as we have shown in \ref{canform}. We can therefore conclude that the momentum variables of each model inherit their Poisson brackets from the Poisson-Lie structure of the dual group, which in turn exhibits Poisson-Lie symmetry in the sense elucidated above.
\section{Conclusions}\label{concl}
Inspired by an existing description of the dynamics of the Isotropic Rigid Rotator on a double Lie group \cite{marmo:articolo1}, we have introduced a new dynamical model which is dual to the standard IRR. To this, we have used the notion of Poisson-Lie groups and Drinfel'd double for understanding the duality between the carrier spaces of the two models. Specifically, we have used the Drinfel'd double of the group $SU(2)$ as the configuration space for the dynamics of a generalized model, with doubled degrees of freedom. This model exhibits non-Abelian duality and is an ideal arena to analyze in a simple context the meaning to physics of generalized and doubled geometry structures.
Moreover, we have shown that, from the generalized action, the usual description with half the degrees of freedom, can be recovered by gauging one of its symmetries.
The simple model of the IRR is especially interesting as a toy model for field theories with non-trivial target spaces such as Principal Chiral Models. In their original formulation \cite{gursey} these are nonlinear sigma models with the principal homogeneous space of the Lie group $SU(N)$ as its target manifold, where $N$ is the number of quark flavors. An action functional, which is formally analogous to the one introduced for the IRR can be written in the form
\be
S=\frac{1}{2}\int_{\mathbb{R}^2}Tr(g^{-1}\mathrm{d}g \wedge * g^{-1}\mathrm{d}g),\label{chimoac}
\ee
where trace is understood as the scalar product on the Lie algebra $\mathfrak{g}.$ The Hodge operator exchanges the time and space derivatives
\be
* (g^{-1}\mathrm{d}g)= *( \dot Q^i dt+{Q^i}' d\sigma ) e_i= ( \dot Q^i d\sigma-{Q^i}'dt ) e_i
\ee
with $ \dot Q^i =\Tr g^{-1}\partial_{t}g e_i$, $ { Q^i}' =\Tr g^{-1}\partial_{\sigma}g e_i$. It describes the dynamics of two dimensional fields $g: \mathbb{R}^{1,1}\rightarrow G$, with
\be
g^{-1}\mathrm{d}g=g^{-1}\partial_{t}g \mathrm{d}t+g^{-1}\partial_{\sigma}g \mathrm{d}\sigma
\ee
the Lie algebra valued left-invariant one-forms.
Therefore, the dynamical fields of the model take value in the cotangent bundle of the Lie group, while the canonical formalism is described by a Poisson algebra which takes the form of a semi-direct sum. The analogy with the IRR is thus very strict: the analysis we have performed can be readily generalized, starting from an alternative description of Principal Chiral Models given in refs \cite{rajeev:bosonization, vitale1, vitale2, delduc} (also see \cite{reid, copland} where sigma models are analyzed in the DFT context). We are completing the analysis and the results will be detailed in a forthcoming paper \cite{MPV18_2}.
\noindent{\bf Acknowledgements}
P. V. acknowledges support by COST (European Cooperation in Science and Technology) in the framework of COST Action MP1405 QSPACE.
\begin{thebibliography}{9}
\bibitem{Klim} Klim\v{c}\'{i}k C. and \v{S}evera P., \emph{Dual Non-Abelian Duality and the Drinfel'd Double}, Phys. Lett. {\bf B351} (1995), no. 4, 455-462, {\tt [hep-th/9502122]} ; Klim\v{c}ik and P. Severa, \emph{Poisson-Lie T-Duality and Loop Groups of Drinfel'd doubles}, Phys. Lett. {\bf B372} (1996) 65 [hep-th/9512040].
\bibitem{Klim2} Klim\v{c}\'{i}k C., \emph{Poisson-Lie T-duality}, Nucl. Phys. Proc. Suppl. {\bf 46} (1996) 116 {\tt [hep-th/9509095]}.
\bibitem{lledo} Lled'{o} M. A. and Varadarajan V. C., \emph{SU(2) Poisson-Lie T duality} Lett. Math.Phys. {\bf 45} (1998) 247.
\bibitem{hitchin1}
Hitchin N., \emph{Generalized Calabi-Yau Manifolds}, Q. J. Math. 54 (2003) 281-308, {\tt [math/0209099]}.
\bibitem{hitchin2}
Hitchin N., \emph{Lectures on Generalized Geometry}, {\tt [arXiv:1008.0973 [math.DG]]}.
\bibitem{gualtieri:tesi}
Gualtieri M., \emph{Generalized Complex Geometry}, PhD Thesis (2004), {\tt [math/0703298 [math.DG]]}.
\bibitem{hulled}
Hull C., Reid-Edwards R.A. (2009), \emph{Non-geometric backgrounds, doubled geometry and generalised T-duality}, JHEP {\bf09} (2009) 014.
%doi:10.1016/0550-3213(84)90453-X
%%CITATION = doi:10.1016/0550-3213(84)90453-X;%%
\bibitem{HZ}
Hull C. and Zwiebach B., \emph{Double Field Theory}, JHEP {\bf 0909} (2009) 099, {\tt [arXiv:0904.4664 [hep-th]]}. \\
Hull C. M. and Zwiebach B., \emph{The Gauge Algebra of Double Field Theory and Courant Brackets}, JHEP {\bf 0909} (2009) 090, {\tt [arXiv:0908.1792 [hep-th]]} .
\bibitem{drinfel'd}
Drinfel'd V. G., \emph{Hamiltonian structures of Lie groups, Lie bialgebras and the geometric meaning of the classical Yang-Baxter equation}, Sov. Math. Dokl. 27, (1983) 68-71 ;
Drinfel'd V. G., \emph{Quantum groups}, Proceedings of the International Congress of Mathematicians (Berkeley, Calif., 1986), American Mathematical Society, Providence (1987) pp. 798-820.
\bibitem{semenov}
Semenov-Tian-Shansky M.~A., \emph{Poisson Lie groups, quantum duality principle, and the quantum double},
Theor.\ Math.\ Phys.\ {\bf 93} (1992) 1292
[Teor.\ Mat.\ Fiz.\ {\bf 93N2} (1992) 302 ], {\tt [hep-th/9304042]}.
\bibitem{MPV18} Marotta V. E., Pezzella F., and Vitale P., \emph{Doubling, T-Duality and Generalized Geometry: a Simple Model} JHEP (2018) {\tt [arXiv:1804.00744 [hep-th]]}.
%%CITATION = ARXIV:1804.00744;%%
\bibitem{MPV18_2}
Marotta V. E., Pezzella F., and Vitale P., {\em in preparation}.
\bibitem{marmo:articolo1}
Marmo G., Simoni A., and Stern A., \emph{Poisson-Lie group symmetries for the isotropic rotor}, Int. J. Mod. Phys. {\bf A10} (1995) 99, {\tt [hep-th/9310145]}.
%%CITATION = doi:10.1142/S0217751X9500005X;%%
\bibitem{MI98} Marmo G. and Ibort A.,
\emph{A new look at completely integrable systems and double Lie groups},
Contemp.\ Math.\ {\bf 219} (1998) 159.
%%CITATION = CTMAE,219,159;%%
%\bibitem{porrati} Giveon A., Porrati M., and Rabinovici E., \emph{Target Space Duality in String Theory}, Phys. Rept. {\bf 244} (1994) 77.
%\bibitem{alvarez} Alvarez E., Alvarez-Gaum\'e, L. and Lozano Y., Nucl. Phys. (Proc. Supp.) 41 (1995) 1
\bibitem{Tseytlin} Tseytlin A. A., \emph{Duality Symmetric Formulation of String World Sheet Dynamics}, Phys. Lett. {\bf B242} (1990) 163 ; Tseytlin A. A., \emph{Duality Symmetric Closed String Theory and Interacting Chiral Scalars}, Nucl. Phys. {\bf B350} (1991) 395.
\bibitem{Duff} Duff M. J, \emph{Duality Rotations in String Theory}, Nucl. Phys. {\bf B335} (1990) 610 .
\bibitem{siegel} Siegel W., \emph{Superspace duality in low-energy superstrings,} Phys. Rev. {\bf D48}, 2826 (1993) {\tt [hep-th/9305073]} ; Siegel W., \emph{Two vierbein formalism for string inspired axionic gravity}, Phys. Rev. {\bf D47} (1993) 5453 {\tt [hep-th/9302036}]; Siegel W. \emph{Manifest duality in low-energy superstrings}, in ``Berkeley 1993, Proceedings, Strings '93" 353-363, {\tt [hep-th/9308133]}; Siegel W, \emph{Manifest Lorentz Invariance Sometimes Requires Nonlinearity}, Phys. Rev. {\bf D47} (1993) 5453.
\bibitem{siegel84}
W.~Siegel,
``Manifest Lorentz Invariance Sometimes Requires Nonlinearity,''
Nucl.\ Phys.\ B {\bf 238} (1984) 307.
doi:10.1016/0550-3213(84)90453-X
%%CITATION = doi:10.1016/0550-3213(84)90453-X;%%
\bibitem{Hull} Hull C. M., \emph{A Geometry for non-Geometric String Backgrounds}, JHEP {\bf 0510} (2005) 065.
\bibitem{lledo} Lled'{o} M. A. and Varadarajan V. C., \emph{SU(2) Poisson-Lie T duality} Lett. Math.Phys. {\bf 45} (1998) 247.
\bibitem{sfetsos} Sfetsos K., \emph{Duality-invariant class of two-dimensional field theories}, Nucl. Phys. {\bf B561} (1999) 316-340, {\tt [hep-th/9904188]}.
\bibitem{falceto} Calvo I., Falceto F. and Garcia-Alvarez D., \emph{Topological Poisson sigma models on Poisson Lie groups,}
JHEP {\bf 0310}, 033 (2003) {\tt [hep-th/0307178]}.
\bibitem{hassler}
Blumenhagen R., Hassler F. and L\"ust D.,
\emph{Double Field Theory on Group Manifolds} ,
JHEP {\bf 1502}, 001 (2015)
{\tt [arXiv:1410.6374 [hep-th]]};
%%CITATION = doi:10.1007/JHEP02(2015)001;%%
\\
Blumenhagen R., du Bosque P., Hassler F., L\"ust D., \emph{Generalized Metric Formulation of Double Field Theory on Group Manifolds}, JHEP {\bf 1508} 056 (2015) {\tt [arXiv:1502.02428 [hep-th]]};
%%CITATION = doi:10.1007/JHEP08(2015)056;%%
\\
Hassler F., \emph{Poisson-Lie T-Duality in Double Field Theory} {\tt [arXiv:1707.08624 [hep-th]]}
\bibitem{marmosaletan} Marmo G., Saletan E. J., Simoni A., Vitale B. , \emph{Dynamical Systems: A Differential Geometric Approach to Symmetry and Reduction}, John Wiley and Sons Inc. (1985)
\bibitem{alex} Alekseev~A.~Y. and Malkin~A.~Z., \emph{Symplectic structures associated to Lie-Poisson groups}, Commun.\ Math.\ Phys.\ {\bf 162} (1994) 147, {\tt hep-th/9303038}.
\bibitem{babelon} O. Babelon and D. Bernard, Phys. Lett. B260 (1991) 81; Commun. Math. Phys. 149 (1992) 279; Int. J. of Mod. Phys. A8 (1993) 507.
\bibitem{wein}
Zhang-Ju Liu, Ping Xu, and Weinstein A., \emph{Manin triples for Lie bialgebroids}, J. Differential Geom. 45 (1997), no. 3, 547--574, {\tt dg-ga/9508013v3}.
\bibitem{freidel}
Freidel L., Rudolph F.J., and Svoboda D., \emph{Generalised Kinematics for Double Field Theory}, JHEP {\bf 11} (2017) 175, {\tt [arXiv:1706.07089 [hep-th]]}.
\bibitem{skly} E. K. Sklyanin, \emph{Some algebraic structures connected with the Yang?Baxter equation}, Funct. Analy. Appl. 16 (1982) 263.
\bibitem{gursey} G\"ursey F. , \emph{On the symmetries of strong and weak interactions}, Il Nuovo Cimento. {\bf 16} (2): 230, (1960); F. G\"ursey, \emph{On the structure and parity of weak interaction currents}, Annals of Physics, {\bf 12} 91, (1961).
\bibitem{rajeev:bosonization}
Rajeev S.G., \emph{Non Abelian bosonization without Wess-Zumino terms. 1. New current algebra}, Phys. Lett. {\bf B217} ;
%%CITATION = doi:10.1016/0370-2693(89)91528-1;%%
\\
Rajeev S.G., \emph{Non Abelian bosonization without Wess-Zumino terms. 2.} (1988) UR-1088.
%%CITATION = UR-1088;%%
\bibitem{vitale1}
Rajeev S.G., Sparano G., and Vitale P., \emph{Alternative Canonical Formalism for the Wess-Zumino-Witten Model}, Int. J. Mod. Phys. {\bf A9} (1994) 5469-5488, {\tt [hep-th/9312178]}.
%%CITATION = doi:10.1142/S0217751X94002211;%%
\bibitem{vitale2}
Rajeev S.G., Stern A., and Vitale P., \emph{Integrability of the Wess-Zumino-Witten model as a non-ultralocal theory}, Phys. Lett. {\bf B388} (1996) 769-775, {\tt [hep-th/9602149]}.
%%CITATION = doi:10.1016/S0370-2693(96)01224-5;%%
\bibitem{delduc}
Delduc F., Magro M., and Vicedo B., \emph{Integrable double deformation of the principal chiral model}, Nucl. Phys. {\bf B891} (2105) 312-321, {\tt [arXiv:1410.8066 [hep-th]]}.
\bibitem{reid}
Reid-Edwards R.A. (2010), \emph{Bi-algebras, generalized geometry and T-duality}, {\tt [arXiv:1001.2479 [hep-th]]}.
\bibitem{copland} Copland N. B., \emph{A double sigma model for double field theory}, JHEP {\bf 04} (2012) 575 ; Copland N.B., \emph{Connecting T-duality invariant theories}, Nucl. Phys. {\bf B854} (2012) 044.
% \bibitem{park}Lee K. and Park J.-H., \emph{Covariant action for a string in doubled yet gauged spacetime} Nucl. Phys. {\bf B880} (2014) 134 ; Park J.-H., \emph{Comments on double field theory and diffeomorphisms}, JHEP {\bf 1306} (2013) 098.
%\bibitem{Berman} Berman D. S. and Thompson D. C., \emph{ Duality Symmetric String and M-Theory}, Phys. Rept. {\bf 566} (2014) 1 ; Berman D. S. , Copland N. B., and Thompson D. C. \emph{Background Field Equations for the Duality Symmetric String}, Nucl. Phys. {\bf B791} (2008) 175.
%\bibitem{pezzella} F. Pezzella, \emph{Some Aspects of the T-Duality Symmetric String Sigma Model}, Proceedings, 14th Marcel Grossmann Meeting on Recent Developments in Theoretical and Experimental General Relativity, Astrophysics, and Relativistic Field Theories (MG14) (2016) ; F. Pezzella, \emph{Two Double String Theory Actions: Non-Covariance vs. Covariance} PoS CORFU2014 (2015) 158; De Angelis L., Gionti G. S. J., Marotta R. and Pezzella F., \emph{Comparing Double String Theories}, JHEP {\bf1404}, (2014) 171.
%\bibitem{Bandos} Bandos I., \emph{Superstring in Doubled Superspace} Phys. Lett. {\bf B751} (2105) 408.
%\bibitem{nibbelink} Nibbelink S. G., \emph{A Lorentz invariant doubled world-sheet}, Phys. Rev. {\bf D87} (2013) 4.
%\bibitem{Ma} Ma C.-T., \emph{One-Loop $\beta$-Function of the Double Sigma Model with Constant Background}, JHEP {\bf 1504} (2015) 026.
%\bibitem{siegel} Siegel W., \emph{Superspace duality in low-energy superstrings,} Phys. Rev. {\bf D48}, 2826 (1993) {\tt [hep-th/9305073]} ; Siegel W., \emph{Two vierbein formalism for string inspired axionic gravity}, Phys. Rev. {\bf D47} (1993) 5453 {\tt [hep-th/9302036}]; Siegel W. \emph{Manifest duality in low-energy superstrings}, in ``Berkeley 1993, Proceedings, Strings '93" 353-363, {\tt [hep-th/9308133]}; Siegel W, \emph{Manifest Lorentz Invariance Sometimes Requires Nonlinearity}, Phys. Rev. {\bf D47} (1993) 5453;
% \bibitem{HZ2} Hull C. M. and Zwiebach B., \emph{The Gauge Algebra of Double Field Theory and Courant Brackets}, JHEP {\bf 0909} (2009) 090, {\tt [arXiv:0908.1792 [hep-th]]} .
%\bibitem{AF} Alekseev A.Y. and Faddeev L.D. , \emph{$(T^*G)_t$: a toy model for conformal field theory}, Comm. Math. Phys. {\bf 141} (1991) 413 .
%\bibitem{deser1}
%Deser A., Stasheff J. (2014), \emph{Even symplectic supermanifolds and double field theory}, Commun. Math. Phys. (2015) 339: 1003, {\tt [arXiv:1406.3601 [math-ph]]}.
%\bibitem{deser2}
%Deser A., Saemann C. (2016), \emph{Extended Riemannian Geometry I: Local Double Field Theory}, ITP-UH-22-16, EMPG-16-18, {\tt [arXiv:1611.02772 [hep-th]]}.
%\bibitem{lust} L\"{u}st D. and Osten D., \emph{Generalized fluxes, Yang-Baxter deformations and the O(d,d) structure of non-abelian T-duality},
%{\tt [arXiv:1803.03971 [hep-th]]}.
%%CITATION = ARXIV:1803.03971;%%
% \bibitem{KS18} V.~G.~Kupriyanov and R.~J.~Szabo,
%\emph{Symplectic realisation of electric charge in fields of monopole distributions,'}
%{\tt [arXiv:1803.00405 [hep-th]]}.
%%CITATION = ARXIV:1803.00405;%%
% \bibitem{majid} S. Majid Pacific J. Math. {\bf 141} (1990) 311
%%CITATION = doi:10.1088/1126-6708/2003/10/033;%%
% \bibitem{duflo}
% Rosa L.~ and Vitale P.,
%{\emph On the $\star$-product quantization and the Duflo map in three dimensions},
% Mod.\ Phys.\ Lett.\ A {\bf 27}, 1250207 (2012) {\tt [arXiv:1209.2941 [hep-th]]}.
%%CITATION = doi:10.1142/S0217732312502070;%%
% \bibitem{noi} Carinena J.~F.~, Gracia-Bondia J.~M., Lizzi F., Marmo G., and Vitale P., \emph{Star-product in the presence of a monopole}, Phys.\ Lett. {\bf A374} (2010) 3614, [arXiv:0912.2197 [math-ph]]. %%CITATION = doi:10.1016/j.physleta.2010.06.069;%%
% \bibitem{vitalewallet} Vitale P. and Wallet J.~C., \emph{Noncommutative field theories on $R^3_\lambda$: Toward UV/IR mixing freedom}, JHEP {\bf 1304}, 115 (2013) Addendum: [JHEP {\bf 1503}, 115 (2015)] {\tt [arXiv:1212.5131 [hep-th]]} ; %%CITATION = doi:10.1007/JHEP04(2013)115, 10.1007/JHEP03(2015)115;%%
% \\
% G\'er\'e A., Vitale P. and Wallet J.C., \emph{Quantum gauge theories on noncommutative three-dimensional space},
% Phys.\ Rev.\ D {\bf 90}, no. 4, 045019 (2014)
% {\tt [arXiv:1312.6145 [hep-th]]}; %%CITATION = doi:10.1103/PhysRevD.90.045019;%%
% \\
% Vitale P.,
%\emph{Noncommutative field theory on $\mathbb{R}^3_\lambda$}', Fortsch.\ Phys.\ {\bf 62}, 825 (2014) {\tt [arXiv:1406.1372 [hep-th]]}.
% %%CITATION = doi:10.1002/prop.201400037;%%
% \bibitem{kupr} Kupriyanov V.~G. and Vitale P., {\emph Noncommutative $ {\mathrm{\mathbb{R}}}^d $ via closed star product}, JHEP {\bf 1508}, 024 (2015), {\tt [arXiv:1502.06544 [hep-th]]}.
% %%CITATION = doi:10.1007/JHEP08(2015)024;%%
% \bibitem{wallet} Juric T., Poulain T., and Wallet J.~C., \emph{Closed star product on noncommutative $\mathbb{R}^3$ and scalar field dynamics}, JHEP {\bf 1605}, 146 (2016) {\tt [arXiv:1603.09122 [hep-th]]} \, ; \,
% %%CITATION = doi:10.1007/JHEP05(2016)146;%%
% \\
% G\'er\'e A., Juric T. and Wallet J.~C.,
% \emph{Noncommutative gauge theories on $ {\mathrm{\mathbb{R}}}_{\lambda}^3 $ : perturbatively finite models},
% JHEP {\bf 1512}, 045 (2015), {\tt [arXiv:1507.08086 [hep-th]]}.
% \bibitem{zoupanos} Chatzistavrakidis A., Jonke L., Jurman D., Manolakos G., Manousselis P., and Zoupanos G.,
%\emph{Noncommutative Gauge Theory and Gravity in Three Dimensions},
%{\tt [arXiv:1802.07550 [hep-th]]}.
%\bibitem{gualtieri:tesi}
%Gualtieri M., \emph{Generalized Complex Geometry}, PhD Thesis (2004), {\tt [math/0703298 [math.DG]]}.
%\bibitem{Hohm}
%Hohm O., Hull C., and Zwiebach B., \emph{Generalized Metric Formulation of Double Field Theory}, JHEP {\bf1008} (2010) 008, {\tt arXiv:1006.4823v2 [hep-th]}
\end{thebibliography}
\end{document}