Flat minimal tori and Lu's second-gap conjecture

Abstract

Lu conjectured that, for each dimension and codimension, there exists a positive gap above the first pinching value for the quantity $S+{\lambda }_2$ on closed minimal submanifolds of the unit sphere, where $S$ is the squared norm of the second fundamental form and ${\lambda }_2$ is the second eigenvalue of Lu's fundamental matrix. We disprove this second-gap conjecture for minimal surfaces in every codimension $q\ge 3$. More precisely, in every odd codimension $q\ge 3$ we construct linearly full closed embedded flat minimal tori with $S\equiv 2$ for which the constant values of $S+{\lambda }_2$ are dense in $(2,3)$. Thus the first pinching value $2$ can be approached from above by closed embedded minimal surfaces, and no uniform second gap exists in codimension at least three. Together with the known positive results in codimensions one and two, our examples complete the codimension picture for Lu's second-gap problem for minimal surfaces: the conjecture holds precisely in codimensions $1$ and $2$,

32 pages, any comments are welcome!