A characterization of the Clifford torus (Q4226293)

The authors obtain sufficient conditions for a closed minimal hypersurface \(M^n\) of a unit sphere to be a Clifford torus. The main point is to consider a pinching estimate \(n\leq S\leq n+\varepsilon(n)\) of the length \(S\) of the second fundamental form of the surface \(M\). A natural and interesting problem, which has not been solved completely yet, is to prove that if there exists \(\varepsilon(n)>0\) in the previous inequality, then \(M^n\) should be a Clifford torus. This can be regarded as a generalization of the well-known result due to \textit{H. B. Lawson} [Ann. Math., II. Ser. 89, 187-197 (1969; Zbl 0174.24901)] and \textit{S. S. Chern, M. P. do Carmo} and \textit{S. Kobayashi} [in: Functional analysis and related fields, Chicago 1968, 59-75 (1970; Zbl 0216.44001)] on the characterization of Clifford tori as the only \(n\)-dimensional minimal surfaces which have constant length \(S=n\) in the unit \((n+1)\)-sphere. The authors prove that the constant \(\varepsilon(n)=\frac{14(n+4)}{9n+30}\) and an additional assumption that the Ricci curvature \(\text{Ric} (M)\geq \frac{n}{2}\) will imply that \(S=n\). As a consequence, \(M\) is a Clifford torus. It is very interesting that, in the case \(n\leq 5\), the authors improve the previous result. Namely, they show that the additional assumption on the Ricci curvature may be omitted and the rigidity result is true if \(\varepsilon(3)=\frac{42}{85}\), \(\varepsilon(4)=\frac{8}{31}\) and \(\varepsilon(5)=\frac{3(21-5\sqrt{17})}{28+3\sqrt{17}}\).