On framed cobordism classes of classical Lie groups (Q1425463)

A compact connected Lie group \(G\) with left invariant framing \(L\) defines via the Thom-Pontrjagin construction an element \([G,L]\) in the stable homotopy groups of spheres. In this paper the problem is considered of determining whether or not this is zero when \(G\) is a classical Lie group. Such a problem has been studied by many authors. This work is especially motivated by the result by \textit{E. Ossa} [Topology 21, 315--323 (1982; Zbl 0491.55008)] stating that \(24[G,L]=0\). The result proved here is: \([SO(2n+1), L]_{(3)}=0\) and \([Sp(n),L]_{(3)}=0\) for any \(n\geq 3\), \(n\neq 5\), 7, 11 where the subscript (3) indicates the 3-component of \([G,L]\). One previously knows that \([SO(2n),L]_{(3)}=0\) \((n\geq 1)\) and \([SU(n),L]_{(3)}=0\) \((n\geq 3)\) and also \([SO(3),L]\) and \([SO(5), L]\) generate the 3-components of the 3- and 10-dimensional stable homotopy groups of spheres, respectively. So it turns out that only the six ones above remain unknown in the classical case. The proof uses a theorem of \textit{J. C. Becker} and \textit{R. Schultz} [Fixed point indices and left invariant framings, Springer Lecture Notes 657, 1--31 (1977; Zbl 0391.55014)] and an idea of E. Ossa for the proof of the result mentioned above. Also, the author's proof requires a formula of \textit{B. Steer} [Topology 15, 383--393 (1976; Zbl 0336.57011)] concerning framings twisted by representations.