Some loci of rational cubic fourfolds (Q1741817)

The rationality of the smooth cubic fourfold is a long standing open problem in the field. The very general element in the moduli space \(\mathcal{C}\) of smooth cubic hypersurfaces in \(\mathbb{P}^5\) is expected to be irrational, however no cubic fourfold has been proven to be not rational yet. The rational cubic fourfolds are conjectured to be exactly the ones parametrized by the infinite countable collection of divisors \(\mathcal{C}_d\) in \(\mathcal{C}\), introduced via Hodge Theory by \textit{B. Hassett} [Compos. Math. 120, No. 1, 1--23 (2000; Zbl 0956.14031)], for \(d\) even and not divisible by \(4\), \(9\) or any prime number congruent to \(5\) modulo \(6\). In the paper under review, the authors study the divisor \(\mathcal{C}_{14}\) proving that all the varieties in the divisor are rational. The general smooth cubic hypersurface parametrized by \(\mathcal{C}_{14}\) contains a smooth quartic rational normal scroll, hence it is rational. The rationality of any element in \(\mathcal{C}_{14}\) is proven through a deep analysis of the possible degenerations of quartic rational normal scrolls contained in the cubic fourfolds. Namely, they prove that all the limits of quartic rational normal scrolls in \(\mathbb{P}^5\) contained in a smooth cubic hypersurface are surfaces with one apparent double point. It needs to be mentioned that, at the time of this review, the result follows from the fact that rationality specializes in smooth families of arbitrary dimension, as proved by \textit{M. Kontsevich} and \textit{Y. Tschinkel} [Invent. Math. 217, No. 2, 415--432 (2019; Zbl 1420.14030)]. However, when the paper under review was written this strong result was unavailable and rationality was known to specialize only in smooth families of threefolds [\textit{T. De Fernex} and \textit{D. Fusi}, Rend. Circ. Mat. Palermo (2) 62, No. 1, 127--135 (2013; Zbl 1264.14019)].