The center of the partition algebra (Q827032)

The paper under review is concerned with the partition algebras \(\mathcal{A}_{2k}(\delta)\) introduced by \textit{P. Martin} [Potts models and related problems in statistical mechanics. Singapore etc.: World Scientific (1991; Zbl 0734.17012); J. Algebra 183, No. 2, 319--358 (1996; Zbl 0863.20009)] and \textit{V. F. R. Jones} [in: Subfactors. Proceedings of the Taniguchi symposium on operator algebras, Kyuzeso, Japan, July 6--10, 1993. 32nd Taniguchi international symposium. Singapore: World Scientific. 259--267 (1994; Zbl 0938.20505)]. These are finite-dimensional associative \(\mathbb{C} \)-algebras depending on a non-negative integer \(k\) and a parameter \(\delta \in \mathbb{C}\). They have a \(\mathbb{C}\)-basis consisting of all partitions of a set with \(2k\) elements. The product of two basis elements is defined via the concatenation of certain diagrams and involves the parameter \(\delta\). The partition algebras are cellular in the sense of \textit{J. J. Graham} and \textit{G. I. Lehrer} [Invent. Math. 123, No. 1, 1--34 (1996; Zbl 0853.20029)]. Their structure and representation theory is closely related to that of the finite symmetric groups. They fit into a chain of \(\mathbb{C}\)-algebras \[ \mathbb{C} = \mathcal{A}_0(\delta) \subseteq \mathcal{A}_1(\delta) \subseteq \mathcal{A}_2(\delta) \subseteq \mathcal{A}_3(\delta) \subseteq \ldots\; . \] For each \(r\), \textit{T. Halverson} and \textit{A. Ram} [Eur. J. Comb. 26, No. 6, 869--921 (2005; Zbl 1112.20010)] have defined a Jucys-Murphy element \(L_r \in \mathcal{A}_r(\delta)\) centralizing \(\mathcal{A}_{r-1}(\delta)\). Here, the author works with the normalized Jucys-Murphy element \(N_r := L_r - \frac{\delta}{2}\). One of the main results of the paper shows that the supersymmetric polynomials in \(N_1,\ldots,N_r\) form a central subalgebra \(\mathrm{SS}_r[N_1,\ldots, N_r]\) of \(\mathcal{A}_r(\delta)\). Moreover, \(\mathrm{SS}_{2k}[N_1,\ldots,N_{2k}]\) equals the center of \(\mathcal{A}_{2k}(\delta)\) whenever \(\mathcal{A}_{2k}(\delta)\) is semisimple. It is known that \(\mathcal{A}_{2k}(\delta)\) is semisimple if and only if \(\delta \notin \{0,1,\ldots,2k-2\}\). Now suppose that \(\mathcal{A}_{2k}(\delta)\) is not semisimple. Then every simple \(\mathcal{A}_{2k} (\delta)\)-module determines a central character. Two simple \(\mathcal{A}_{2k}(\delta)\)-modules belong to the same block if their central characters coincide. The author proves that this is the case when the restrictions of the central characters to \(\mathrm{SS}_{2k}[N_1,\ldots,N_{2k}]\) agree.