Higher-dimensional \(BF\) theories in the Batalin-Vilkovisky formalism: The BV action and generalized Wilson loops (Q5953664): Difference between revisions

From MaRDI portal
Set OpenAlex properties.
Importer (talk | contribs)
Changed an Item
 
Property / arXiv ID
 
Property / arXiv ID: math/0010172 / rank
 
Normal rank

Latest revision as of 11:31, 18 April 2024

scientific article; zbMATH DE number 1695584
Language Label Description Also known as
English
Higher-dimensional \(BF\) theories in the Batalin-Vilkovisky formalism: The BV action and generalized Wilson loops
scientific article; zbMATH DE number 1695584

    Statements

    Higher-dimensional \(BF\) theories in the Batalin-Vilkovisky formalism: The BV action and generalized Wilson loops (English)
    0 references
    27 January 2002
    0 references
    Superformalism of the Batalin-Vilkovisky (BV) quantization procedure for \(BF\) theories is presented. The classical \(BF\) action in \({\mathcal S}^{cl}= \int_M (B,F_A)\), where \(M\) is an \(m\)-dimensional manifold, \(A\) a connection 1-form on \(P\), a principal \(G\)-bundle over \(M\), \(F_A\) the curvature of \(A\) and \(B\) is an \((m-2)\)-form of the adjoint type. In this paper, the authors assume: A1. \(M\) is compact and there is a flat connection \(A_0\) of \(P\) such that all cohomology groups \(H^*_{d_{A_0}}(M,\text{ad }P)\) are trivial. A2. \(P\) is trivial. A3. The Lie algebra \({\mathfrak g}\) of \(G\) is endowed with a symmetric, Ad-invariant, nondegenerate bilinear form \(\langle,\rangle\). The authors say A1 is a strong assumption, but if there is an odd-dimensional manifold satisfying A1, then there exists an even-dimensional manifold satisfying A1, and there might exist topological obstructions to the existence of manifolds satisfying A1 (Section 2). The Euler-Lagrange equation of motion for the \(BF\) action is \(F_A= 0\), \(d_AB= 0\). The space of solutions of this equation which satisfy A1 is referred to as on shell. The infinitesimal form of the symmetry of this action is \[ A\mapsto A+\varepsilon d_A c,\quad B\mapsto B+\varepsilon([B, c]+ d_A\tau_1), \] \(c\in \Omega^0(M,\text{ad }P)\), \(\tau^1\in \Omega^{m- 3}(M,\text{ad }P)\). These symmetries are reducible on shell. To consistently fix all the symmetries, the introduction of a hierarchy of ghosts is necessary and we have to resort to the BV formalism which generalizes BRST procedures and works for any \(m\geq 2\) (Section 3). BV quantization procedures for \(BF\) theories are reviewed in Section 4. To present the BV superformalism, superforms \[ \begin{aligned}{\mathbf B} &= \sum^{m-2}_{k=1} (-1)^{{k(k- 1)\over 2}} \tau_k+ B+(- 1)^m a^++ c^+,\\ {\mathbf A} &= (-1)^{m+1} c+ A+ (-1)^m B^++ \sum^{m-2}_{k= 1} (-1)^{{k(k- 1)\over 2}+ m(k+ 1)}\tau^+_k,\end{aligned} \] where \(c^+\) means the antifield of \(c\) and the \(\tau_k\)'s are defined by the rules \[ \begin{aligned} \delta_{\text{BRST}}\tau_1 &= -[\tau_1, c]+ d_A\tau_2,\;\delta_{\text{BRST}}\tau_2= [\tau_2, c],\dots,\\ \delta_{\text{BRST}}\tau_k &= (-1)^k[\tau_k, c]+ d_A\tau_{k+ 1},\;\delta_{\text{BRST}}\tau_{m- 2}= (-1)^m[\tau_{m- 2}, c]\end{aligned} \] are introduced. Fixing a background connection \(A_0\), which satisfies A1, \({\mathbf A}\) is written as \(A_0+{\mathbf a}\), \({\mathbf A}\) is a superconnection. The covariant derivative of \({\mathbf B}\) and the supercurvature are defined by \[ d_{{\mathbf A}}= d_{A_0}{\mathbf B}+ [[{\mathbf a};{\mathbf B}]],\quad{\mathbf F}_{{\mathbf A}}= d_{A_0}{\mathbf a}+ \textstyle{{1\over 2}}[[{\mathbf a};{\mathbf a}]]. \] Here the bracket is defined by using the dot product which is explained in Appendix B. In the rest, the algebra of functionals is restricted to only those functionals that depend on linear combinations of \({\mathbf A}\) and \({\mathbf B}\) which is denoted by \({\mathcal S}_{{\mathbf A},{\mathbf B}}\). Values of functionals are either a commutative algebra \({\mathfrak A}\), usually \(\mathbb{R}\) or \(\mathbb{C}\), or a de Rham complex \(\Omega^*(N; E)\), usually \(N\) is \(LM\). After defining superfunctional derivatives, super BV antibracket, by which \({\mathcal S}_{{\mathbf A},{\mathbf B}}\) becomes a Gerstenhaber algebra [\textit{M. Gerstenhaber}, The cohomology structure of an associative ring, Ann. Math. (2) 78, 267-288 (1962; Zbl 0131.27302), On the deformation of rings and algebras, Ann. Math. (2) 79, 59-103 (1964; Zbl 0123.03101)], and the super BV Laplacian is defined. Then the super BV operator is defined which leads to definitions of flat observables and flat invariant observables (Definition 5.7). After these preparations, the BV action for super \(BF\) theory is claimed to be \[ F{\mathcal S}= \int_M\langle\langle{\mathbf B};{\mathbf F}_{{\mathbf A}}\rangle\rangle. \] To justify this claim, it is shown that this action satisfies the master equation of the super \(BF\) theory (Theorem 6.5). Ordinary \(BF\) theory takes \(B\) to be the adjoint type, while taking \(B\) to be the coadjoint type, canonical \(BF\) theory is introduced in subsection 6.3. Action of canonical super \(BF\) theory is also claimed to have the same form and shown to satisfy the master equation. The rest of Section 6 deals with gauge fixing and superpropagators (of ordinary \(BF\) theory). A superpropagator \(\eta\) is a distributional \((m-1)\)-form on \(M\times M\) with values in \(\text{ad }P\boxtimes \text{ad }P\). It can be computed by generalizing the Axelrod and Singer construction [\textit{S. Axelrod} and \textit{I. M. Singer}, Chern-Simons perturbation theory, Proc. of the XXth DGM Conf., 3-45 (1992; Zbl 0313.53051), Chern-Simons perturbation theory. II, J. Differ. Geom. 39, 173-213 (1994; Zbl 0826.53057)] to higher dimensions. Another possibility is to compute main properties of the superpropagator and then construct a form that satisfies them, generalizing the construction of \textit{R. Bott} and \textit{A. S. Cattaneo} [Integral invariants of 3-manifolds. II, J. Differ. Geom. 48, 91-133 (1998; Zbl 0953.57008)]. The properties are \[ T^*\eta= (-1)^m\eta,\quad \iota^*\eta= \vartheta\otimes \pi^{\vartheta*}\phi+ \pi^{\vartheta*}\beta, \] where \(T\) is the automorphism of \(\text{ad }P\boxtimes \text{ad }P\) exchanging points and fibers, \(\vartheta\) is the global angular form of the spherical normal bundle of the diagonal of \(M\times M\) and \(\beta\in \Omega^{m- 1}(\text{Diag, ad }P\otimes \text{ad }P)\). Existence of a covariantly closed \(\eta\) having these properties is shown assuming A1 and A3 (Theorem 6.6). In Section 7, after introducing cosmologically the term \({\mathbf S}_3= 1/6\int_M\langle\langle{\mathbf B};[[{\mathbf B};{\mathbf B}]]\rangle\rangle\), the odd derivation \(\delta= ((S;\;))\), \(S\) a central element of \({\mathcal S}_{{\mathbf A},{\mathbf B}}\), is modified to \(\delta_{\kappa^2}= \delta+ \kappa^2(({\mathbf S}_3;\;))\). Then a variation \({\mathbf C}_\kappa={\mathbf a}+\kappa{\mathbf B}\) of \(A_0\) is introduced and its generalized Wilson loop (Chen's iterated integral [\textit{K. Chen}, Ann. Math. (2) 97, 217-246 (1973; Zbl 0227.58003), explained in Appendix D) is is shown to be \((\delta_{\kappa^2}+d)\)-closed (Theorem 7.2). In Section 8, other loop observables such as \(\int_M \text{Tr }{\mathbf B}^k\) are investigated. This type of observables satisfy a master equation and are \(\Delta_{\text{BV}}\)-exact (Lemma 8.1, 8.5, Theorem 8.10). In Section 9, the last section, after remarking that previous results hold for \(M= \mathbb{R}^m\), assuming superfields decay sufficiently fast at infinity, possibilities to weaken A1 are discussed. Besides Appendices B and D, pushforward of bundles and forms are explained in Appendix A and the universal global angular form is constructed by using a fermionic integral representation in Appendix C.
    0 references
    0 references
    generalization
    0 references
    superconnection
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references