Peskine sixfolds and Debarre–Voisin fourfolds with associated cubic fourfolds
Corey Brooke, Laure Flapan, Sarah Frei, Lisa Marquand · 2025
Read Peskine sixfolds and Debarre–Voisin fourfolds with associated cubic fourfolds by Corey Brooke, Laure Flapan, Sarah Frei, Lisa Marquand (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-18339.txt.
Introduction
Some shapes in algebraic geometry come in linked families, where one shape quietly defines another. Corey Brooke, Laure Flapan, Sarah Frei, and Lisa Marquand study such links for Peskine sixfolds in this 2025 paper. These six-dimensional shapes can pair with K3 surfaces and with cubic fourfolds, shapes cut out by a degree-three equation. The team works out when such pairings can occur.
They derive number-based conditions that a pairing must meet. In the first case where a cubic fourfold appears, at discriminant 24, they name the cubic in full. In that case they also prove a neat match: the shape formed by all lines on the cubic is the same as the paired Debarre-Voisin fourfold. That fourfold is a hyperkähler shape, a class prized in geometry. The results tighten the map of how these shapes relate.
Catalog record
| CREATOR | Corey Brooke, Laure Flapan, Sarah Frei, Lisa Marquand |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-18339
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Peskine sixfolds and Debarre–Voisin fourfolds with associated cubic fourfolds Corey Brooke, Laure Flapan, Sarah Frei, Lisa Marquand Abstract . We develop the notion of Peskine sixfolds with associated K3 surfaces and cubic fourfolds and work out numerical conditions for when these associations occur. In discriminant 24 24 , the first family for which there is an associated cubic fourfold, we identify the cubic explicitly. Moreover, we prove that in this case the Fano variety of lines of the cubic fourfold is isomorphic to the associated Debarre–Voisin hyperkähler fourfold. 1. Introduction There is a close connection between Fano varieties of K3 type and hyperkähler manifolds of K3 [n] -type, most famously exhibited by a cubic fourfold and its Fano variety of lines [ 4 ] . More recently, there have been numerous similar constructions [ 12 , 9 , 19 , 22 , 21 , 13 , 7 ] . Among these, the Debarre–Voisin hyperkähler fourfold introduced in [ 12 ] remains somewhat mysterious. We recall the definition: given a 10 -dimensional complex vector space V_{10} and a trivector \sigma\in\bigwedge^{3}V_{10} , the Debarre–Voisin manifold is defined by the degeneracy condition X_{6}^{\sigma}=\{[V_{6}]\in\mathrm{Gr}(6,V_{10})\;|\;\sigma|_{V_{6}}=0\}. When \sigma is general, this X_{6}^{\sigma} is a hyperkähler fourfold. Initially, Debarre–Voisin observed that X_{6}^{\sigma} was (rationally) Hodge-theoretically associated to a 20-dimensional hypersurface in \mathrm{Gr}(3,V_{10}) . Later, an integral Hodge-theoretic association between X_{6}^{\sigma} and the so-called Peskine sixfold X_{1}^{\sigma}=\{[V_{1}]\in\mathbb{P}(V_{10})\;|\;\mathrm{rank}(\sigma(V_{1},-,-))\leq 6\}, a Fano variety of K3 type, was described [ 15 , 6 ] . Here, we investigate divisors in the moduli space of trivectors \sigma for which X_{6}^{\sigma} and X_{1}^{\sigma} acquire Hodge-theoretic associations to K3 surfaces or cubic fourfolds; when such associations appear, it is natural to expect the birational geometry of X_{6}^{\sigma} and of X_{1}^{\sigma} to be richer. In the case of cubic fourfolds for instance, Hassett divisors \mathcal{C}_{d} in the moduli space of cubic fourfolds parametrize those cubic fourfolds containing a surface not homologous to a complete intersection, where d is the discriminant of the saturated lattice spanned by the class of the surface and the square of the hyperplane class. Hassett described the values of d for which the cubic fourfolds Y\in\mathcal{C}_{d} have associated K3 surfaces, meaning that the transcendental lattice T_{Y}\subset H^{4}(Y,\mathbb{Z}) is Hodge isometric to T_{S}(-1) for some K3 surface S [ 16 ] . For such a Y\in\mathcal{C}_{d} , the hyperkähler fourfold given by the Fano variety of lines F(Y) on Y is always birational to a moduli space of stable sheaves on S [ 1 ] . Further, in this case, the cubic fourfold Y is expected to be rational [ 16 , 20 , 1 , 3 ] . Here we study Peskine sixfolds X_{1}^{\sigma} and their corresponding Debarre–Voisin hyperkähler fourfolds X_{6}^{\sigma} in analogy with the case of cubic fourfolds Y and their corresponding hyperkähler fourfolds F(Y) . Towards this end, we study divisors \mathcal{P}_{d} in the moduli space of trivectors \sigma\in\bigwedge^{3}V_{10}^{\vee} yielding marked Peskine sixfolds X_{1}^{\sigma} or marked Debarre–Voisin fourfolds X_{6}^{\sigma} of discriminant d . We formalize the notion of such \sigma\in\mathcal{P}_{d} having an associated K3 surface or an associated cubic fourfold and establish the following. Theorem 1.1 . Let \sigma\in\bigwedge^{3}V_{10}^{\vee} be a very general trivector of discriminant d. Then: (1) \sigma has an associated K3 surface if and only if d satisfies the numerical conditions of Proposition 4.4 , and (2) \sigma has an associated cubic fourfold Y\in\mathcal{C}_{d} if and only if d satisfies the numerical conditions of Proposition 4.7 . Along the way to proving Theorem 1.1 , we determine orbit representatives for the possible markings for smooth special Peskine varieties, see Proposition 3.7 . The first divisor \mathcal{P}_{d} parametrizing trivectors with an associated cubic fourfold occurs when d=24 . When the trivector \sigma is general in this divisor, the Peskine X_{1}^{\sigma} has exactly one isolated singular point. More specifically, there is a unique flag W_{1}\subset W_{6}\subset V_{10} with \sigma(W_{1},W_{6},V_{10})=0 , and X_{1}^{\sigma} is singular at \mathbb{P}(W_{1}) [ 6 ] . Further, Benedetti–Song show that such a Peskine variety X_{1}^{\sigma} has an associated twisted K3 surface (S,\alpha) with \alpha\in\mathrm{Br}(S) . A main result of [ 5 ] is that when the Brauer class on this K3 surface S vanishes, the Peskine variety X_{1}^{\sigma} is rational. More recently, [ 17 ] studied the case of special cubic fourfolds of discriminant d=24 , and showed that the obstruction to rationality is again governed by a twisted K3 surface. Thus, it is natural to ask whether there is a geometric relationship between a Peskine X_{1}^{\sigma} with \sigma\in\mathcal{P}_{24} and its associated cubic fourfold Y\in\mathcal{C}_{24}. Theorem 1.2 . Let \sigma\in\bigwedge^{3}V_{10}^{\vee} be a very general trivector of discriminant 24. Then: (1) Y:=\mathbb{P}(W_{6})\cap X_{1}^{\sigma} is a smooth cubic fourfold of discriminant 24 , (2) the Fano variety of lines F(Y) is isomorphic to the Debarre–Voisin variety X_{6}^{\sigma} , and (3) Y is Hodge-theoretically associated to \sigma . Outline We recall the definitions of both the Peskine varieties and Debarre–Voisin varieties in Section 2 , along with relevant results on the period maps, moduli spaces and cohomology. In Section 3 , we develop the notion of special trivectors and special Peskine varieties. In Section 4 , we introduce the notion of associated K3 surfaces and cubic fourfolds for special trivectors of discriminant d . We prove that when both are smooth, our definitions coincide with the Hodge-theoretic association analogous to the case of special cubic fourfolds [ 16 ] . We prove Theorem 1.1 in Propositions 4.4 and 4.7 . In Section 5 , we study special trivectors of discriminant 24 more closely, proving Theorem 1.2 . Finally, in Appendix A we provide an explicit example of a trivector with its associated Peskine variety to support claims made in Section 5 . 2. Preliminaries Fixing a ten dimensional complex vector space V_{10} , a trivector \sigma\in\bigwedge^{3}V_{10}^{\vee} defines the following varieties, given by degeneracy conditions: • Peskine variety: X_{1}^{\sigma}:=\{[V_{1}]\in\mathbb{P}(V_{10})\mid\mathrm{rank}(\sigma(V_{1},-,-))\leq 6\}. The variety X_{1}^{\sigma} has expected dimension 6 and is in general smooth. • Debarre–Voisin variety: X_{6}^{\sigma}:=\{[V_{6}]\in\mathrm{Gr}(6,V_{10})\mid\sigma|_{V_{6}}=0\}. This has expected dimension 4, and for general \sigma is a hyperkähler manifold of K3 [2] -type [ 12 ] . The GIT quotient \mathcal{M}_{\sigma}:=\mathbb{P}(\bigwedge^{3}V_{10}^{\vee})\mathbin{/\mkern-6.0mu/}\mathrm{SL}(V_{10}) serves as a moduli space parametrizing both Peskine varieties and Debarre–Voisin varieties. 2.1. The period map Letting \sigma be general so that the variety X_{6}^{\sigma} is indeed a hyperkähler fourfold of K3 [2] -type, the lattice (H^{2}(X_{6}^{\sigma},\mathbb{Z}),q) is isomorphic to \mathbb{L} , where \mathbb{L}:=U^{3}\oplus E_{8}^{2}\oplus\langle-2\rangle. There is a polarization H\in\mathrm{NS}(X_{6}^{\sigma}) satisfying q(H)=22 and q(H,x)\in 2\mathbb{Z} for all x\in H^{2}(X_{6}^{\sigma},\mathbb{Z}) [ 12 , Theorem 1.2] . We denote by \mathcal{M}_{22}^{(2)} the coarse moduli space of K3 [2] -type manifolds with a polarization of square 22 and divisibility 2. Recall that we define the divisibility of v\in L for a lattice L to be the minimal generator of the ideal v\cdot L\subset\mathbb{Z} ; we denote this by \mathrm{div}_{L}(v). There is a unique orbit of primitive elements h\in\mathbb{L} with q_{\mathbb{L}}(h)=22 and \mathrm{div}_{\mathbb{L}}(h)=2 ; we fix such an h . For any X_{6}^{\sigma} , we fix a labeling \phi:H^{2}(X_{6}^{\sigma},\mathbb{Z})\overset{\sim}{\longrightarrow}\mathbb{L} such that \phi(H)=h , yielding a period map \mathscr{P}:\mathcal{M}_{\sigma}\dashrightarrow\mathcal{M}_{22}^{(2)}\rightarrow\mathbb{D}/\Gamma,\hskip 28.45274pt\mathscr{P}([\sigma])=[\phi(H^{2,0}(X_{6}^{\sigma}))], where \mathbb{D}:=\{x\in\mathbb{P}(\mathbb{L}\otimes\mathbb{C})\mid x^{2}=x\cdot h=0,x\cdot\bar{x}>0\} and \Gamma=\mathrm{O}(\mathbb{L},h) . Note that the map \mathcal{M}_{\sigma}\dashrightarrow\mathcal{M}_{22}^{(2)} is birational [ 28 , Theorem 1.8] . Further, the period map \mathscr{P} is resolved by considering the Baily-Borel projective