Solomon-Terao algebra of hyperplane arrangements (Q2302616)

In the paper under review the authors introduce and study a new algebra that can be associated with hyperplane arrangements, the so-called Solomon-Terao algebra. Let us present an outline of the construction, and afterwards the main geometric results related to this object. Let \(\mathbb{K}\) be an algebraically closed field, \(V = \mathbb{K}^{\ell}\), and denote by \(S = \mathbb{K}[x_{1}, \dots, x_{\ell}]\). The \(\mathbb{K}\)-linear \(S\)-derivation module \(\mathrm{Der}(S)\) is a graded free module of rank \(\ell\) defined by \[\mathrm{Der}(S) := \bigoplus_{i=1}^{\ell} S \partial_{x_{i}}.\] We will consider also \(\mathrm{Der}^{p}(S):= \bigwedge^{p} \mathrm{Der}(S)\) for some \(p\geq 0\). Let \(\mathcal{A}\) be a hyperplane arrangement in \(V\) and denote by \(Q(\mathcal{A}) := \prod_{H \in \mathcal{A}} \alpha_{H}\) the defining polynomial, where \(\alpha_{H} \in S_{1}\) such that \(\mathrm{ker}(\alpha_{H}) = H\). Now we can define the logarithmic derivation modules \(D^{p}(\mathcal{A})\) for \(\mathcal{A}\) as follows: \[\mathcal{D}^{p}(\mathcal{A}) = \{ \theta \in \mathrm{Der}^{p}(S) : \theta(\alpha_{H},f_{2}, \dots, f_{p}) \in S\alpha_{H}: \, \forall \, H \in \mathcal{A}, \quad \forall \, f_{2}, \dots, f_{p} \in S\}.\] The Solomon-Terao polynomial \(\Psi(\mathcal{A};x,t)\) is defined as \[\Psi(\mathcal{A};x,t) :=t^{\ell}\sum_{p=0}^{\ell} \mathrm{Hilb}(\mathcal{D}^{p}(\mathcal{A});x) \bigg(\frac{1-x}{t}-1 \bigg)^{p},\] where \(\mathrm{Hilb}(* ;x)\) denotes the Hilbert series. Let now \(d\) be a non-negative integer and \(S_{d}\) is the set of all homogeneous poynomials of degree \(d\). We fix \(\eta \in S_{d}\) and we define the boundary map \[\partial_{p}:\mathcal{D}^{p}(\mathcal{A}) \rightarrow \mathcal{D}^{p-1}(\mathcal{A})\] by \[\partial_{p}(\theta)(f_{2}, \dots, f_{p}) = \theta(\eta,f_{2}, \dots, f_{p})\] for all \(f_{2}, \dots, f_{p} \in S\). We call the complex \((\mathcal{D}^{*}(\mathcal{A}),\partial_{*})\) the Solomon-Terao complex of degree \(d\) with respect to \(\eta \in S_{d}\). We define the cohomology group \[H^{p}(\mathcal{D}^{*}(\mathcal{A}), \partial_{*}):= \frac{\mathrm{ker}\, \partial_{p}}{\mathrm{Im}\, \partial_{p+1}}.\] One can show that there is a non-empty Zariski open set \(U_{d} = U_{d}(\mathcal{A}) \subset S_{d}\) such that every cohomology of the Solomon-Terao complex with respect to \(\eta \in U_{d}\) is of finite dimension over \(\mathbb{K}\), and if the projective dimension \(\mathrm{pd}_{S}\mathcal{D}^{p}(\mathcal{A}) \leq \ell - p\) for all \(p=1,2, \dots, \ell\) (the authors call arrangements with such a property \textit{tame}), then \(H^{i}(\mathcal{D}^{*}(\mathcal{A}),\partial_{*}) = 0\) for \(i\neq0\). Now we can finally define the Solomon-Terao algebra of an arrangement \(\mathcal{A}\) with respect to \(\eta\) as \(\mathrm{ST}(\mathcal{A},\eta) :=H^{0}(\mathcal{D}^{*}(\mathcal{A}),\partial_{*})\). Moreover, we call \(a(\mathcal{A},\eta) = \{\theta(\eta) \in S \, | \, \theta \in \mathcal{D}(\mathcal{A}) \} = \, \mathrm{Im} \partial_{1}\) the Solomon-Terao ideal of \(\mathcal{A}\) with respect to \(\eta\). The first main result of the paper can be formulated as follows. Theorem A. Let \(\mathcal{A}\) be a tame arrangement and \(\eta \in U_{2}(\mathcal{A})\), then we have \[\mathrm{Hilb}(\mathrm{ST}(\mathcal{A},\eta);x) = \Psi(\mathcal{A};x,1).\] In particular, denoting by \(\pi(\mathcal{A},t)\) the topological Poincaré polynomial of the complement \(M(\mathcal{A}) :=V \setminus \bigcup_{H \in \mathcal{A}} H\), we have \(\mathrm{Hilb}(\mathrm{ST}(\mathcal{A},\eta);1) = \pi(\mathcal{A},1)\), i.e., it coincides with the number of chambers when \(\mathbb{K}=\mathbb{R}\), and with the total Betti number \(M(\mathcal{A})\) if only \(\mathbb{K} = \mathbb{C}\). The second part of the paper is devoted to a special class of tame arrangements, namely free arrangements. Let us recall that \(\mathcal{A}\) is free with exponents \(\mathrm{exp}(\mathcal{A}) = \{d_{1}, \dots, d_{\ell}\}\) if \(\mathcal{D}(\mathcal{A})\) is a free \(S\)-module of rank \(\ell\) with homogeneous basis \(\theta_{1}, \dots, \theta_{\ell}\) such that \(\mathrm{deg} \, \theta_{i} = d_{i}\) for all \(i \in \{1, \dots,\ell\}\) It is known that if \(\mathcal{A}\) is free, then \(\mathcal{D}^{p}(\mathcal{A})\) is also free, so the freeness, as it was pointed out above, implies the tameness. Theorem B. For an arrangement \(\mathcal{A}\) and a non-negative integer \(d\), let \(\eta \in U_{d}\). i) Assume that \(\mathcal{A}\) is free with the exponents \(\mathrm{exp}(\mathcal{A}) = (d_{1}, \dots, d_{\ell})\). Then \(\mathrm{ST}(\mathcal{A},\eta)\) is a complete intersection with \[\mathrm{Hil}(\mathrm{ST}(\mathcal{A},\eta);x) = \prod_{i=1}^{\ell}\left(1+x+ \dots + x^{d_{i}+d-2}\right),\] and hence the socle degree of \(\mathrm{ST}(\mathcal{A},\eta)\) is equal to \(|\mathcal{A}| + \ell(d-2)\). ii) Conversly, if \(\mathrm{ST}(\mathcal{A},\eta)\) is a complete intersection, then \(\mathcal{A}\) is free. In the last section, the authors formulate a plethora of questions that might be interesting to study. One of those can be formulated as follows. Question. Is there a topological meaning of \(\Psi(\mathcal{A};x,t)\) and \(\mathrm{Hilb}(\mathrm{ST}(\mathcal{A},\eta);x)\)?