Simple Lie algebras of small characteristic. III: The toral rank 2 case, (Q5945622)

This paper is the third one in a series of papers aiming at the classification of finite-dimensional simple Lie algebras over an algebraically closed field of prime characteristic \(p>3\). In particular, this would confirm the generalized Kostrikin-Shafarevich conjecture which says that every finite-dimensional simple Lie algebra over an algebraically closed field of prime characteristic \(p>5\) is either classical or of Cartan type. The only exception in characteristic 5 seems to be the family of Melikyan algebras. The strategy of the authors is to generalize the existing classification for \(p>7\) to the cases \(p=5\) and \(p=7\). The final step in the existing classification for \(p>7\) is to prove that any finite-dimensional simple Lie algebra contains a maximal subalgebra which satisfies the conditions in the recognition theorem of \textit{R. L. Wilson} [J. Algebra 40, 418--465 (1976; Zbl 0355.17012)]. In order to construct such a maximal subalgebra one needs a ``very special'' torus (i.e., a so-called non-rigid standard torus having only proper roots) in the derivation algebra of the given simple Lie algebra. The existence of such a torus relies on the classification of simple Lie algebras of absolute toral rank 2. The goal of the paper under review is to extend this classification to characteristic 5 and 7. The proof relies heavily on the methods and terminology of the first two parts [J. Algebra 189, 419--480 (1997; Zbl 0878.17019) resp. J. Algebra 216, 190--301 (1999; Zbl 0939.17016)] of this series. The crucial step in the classification of finite-dimensional restricted simple Lie algebras over an algebraically closed field of prime characteristic \(p>7\) due to \textit{R. E. Block} and \textit{R. L. Wilson} [J. Algebra 114, 115--259 (1988; Zbl 0644.17008)] was adapted to the non-restricted case by the second author [Ann. Math. (2) 130, 643--677 (1989; Zbl 0699.17016)]. The main result of the paper under review enables one to extend the latter to the cases \(p= 5\) and \(p= 7\). This will be presented in the next paper of this series. As a consequence, the classification result of \textit{R. E. Block} and \textit{R. L. Wilson} can be extended without much difficulties to the small characteristics. Moreover, with the exception of the final case [Trans. Am. Math. Soc. 350, 2553--2628 (1998; Zbl 0923.17020)] the classification results of the second author need only minor modifications in order to accommodate to the cases \(p=5\) and \(p=7\). New methods are now available in order to solve the final case and to handle the case where two-sections of Melikyan type occur. Therefore it seems very likely that the main obstacles in the classification of the finite-dimensional simple Lie algebras over an algebraically closed field of prime characteristic \(p>3\) are removed by proving the main result of the paper under review.