Non-Abelian geometry (Q1396578)

Non-Abelian geometry means relating spatial noncommutativity and the non-Abelian nature of multiple D-branes. If the spatial noncommutativity and non-Abelian nature of multiple D-branes are decoupled, the algebra of matrix valued fields is thought to be the tensor product of a matrix algebra and the Moyal-Weyl deformation. So to relate them, tensor the product must be deformed. In this paper, applying lattice string quantum mechanics (LSQM), such deformation is constructed, when multiple D-branes are configurated in a flat but varying \(B\) field background, or in the presence of non-Abelian gauge field background. In the single D-brane case, the Hilbert space for the LSQM is the tensor product of the Hilbert spaces \({\mathcal H}_1\) and \({\mathcal H}_2\) of two ends. The authors introduce a new product in this space by \[ (|\alpha\rangle\otimes \langle\beta|)* (|\theta\rangle\otimes \langle\rho|)\propto (\langle\beta\|\theta\rangle)(|\alpha\rangle\otimes \langle\rho|). \] In terms of wave functions, this product is shown to be \[ \begin{aligned} \Psi*\Phi)(x) &= \exp\Biggl({i\over 2} {\partial\over\partial x^{\prime\mu}} \Omega^{\mu\nu} {\partial\over\partial x^{\prime\prime\nu}}\Biggr) \Psi(x')|\Phi(x'')|^{x'= x''= x},\\ \Omega &= -(2\pi\alpha')^2 G^{-1} BG^{-1}(1- (2\pi\alpha')^2 BG^{-1} BG^{-1})^{-1},\end{aligned} \] where \(G\) is the spacetime metric and \(B\) is the \(B\) field. Then the deformed product for the multiple D-branes configured in a flat but varying \(B\) field background case is computed and its associativity is shown (Sect. 2). In Sect. 3, the SU(2) case, that is the noncommutative parameter on brane 1 is \(\Omega\) but that on the brane 2 is \(-\Omega\), is treated and the product becomes \[ \begin{aligned} (\Psi\times\Phi) &\equiv \Psi(e^a, M^A)\exp\Biggl(i {\partial\over\partial M^A} \widehat\Omega^{Aa}{\partial\over\partial M^a}\Biggr) \Phi(M^a, e^A)|^{M^A= 0= M^a},\\ \widehat\Omega^{Aa} & = J^{Aa}\sigma_3,\quad J=\begin{pmatrix} 0 & \mathbb{I}\\ -\mathbb{I} & 0\end{pmatrix},\end{aligned} \] and the authors say that this formula can be extended to the \(N\) branes case. Then the structure of the deformed algebra is discussed and it is shown that this algebra cannot be obtained by tensoring the subalgebra of constant matrices (it is the matrix algebra by the \(\times\) product) with some other algebra. The authors note that in the usual case of \(N\times N\) matrix fields a commutative or U(1) deformed space can be interpreted as the algebra on the direct product of the continuous space with the appropriate discrete torus. However, with non-Abelian deformation considered in this paper, the discrete torus and the continuous space lose their independent identities and separate meanings. The conventional picture has to be replaced by a total space that intertwines the discrete non-Abelian degrees of freedom of the Yang-Mills theory with that of the continuous noncommutative space. This is the precise meaning of non-Abelian geometry. The situation of multiple parameters appears in connection with Taub-NUT geometry. This is discussed in Sect. 4 and it is shown how a whole class of Lorentz non-invariant theories governed by the \(B\) field dynamics can be studied in an unified way. Sect. 5, the last Section, discusses possible supergravity background for the analysis presented in the earlier sections. Then the shift \(\nu\) in the mode number due to the presence of different gauge fluxes at the boundary is computed and show \[ \nu= {2\over\pi} \tan^{-1}F, \] in the \(\text{SU}(2)\)-case. So \(\nu\to 1\) when \(F\to\infty\) in this they case.