Divisibility of the Dirac magnetic monopole as a two-vector bundle over the three-sphere (Q2517774)

This is an intriguing paper. A (non-trivial) magnetic monopole \(\mu\) can be interpreted as a gerbe over the three sphere \(S^3\). Gerbes in turn are 2-vector spaces of rank one, and hence the magnetic monopole \(\mu\) defines an element in the \(K\)-theory group of 2-vector spaces over \(S^3\). The authors prove that in the latter group, there are classes \(\nu\) and \(\theta\) such that \(\mu+\nu=2\theta,\) where \(24\nu = 0\) and \(\theta\) is of infinite order. In particular, ignoring torsion, \(\mu\) is divisible by two. To prove their theorem the authors need to study the effect on the third homotopy groups of a certain map \(K(\mathbb Z, 3)\to K(ku)\) that interprets gerbes as 2-vector bundles; here \(K(\mathbb Z, 3)\) is an Eilenberg-MacLane space and \(K(ku)\) denotes the \(K\)-theory of the connected complex \(K\)-theory ring spectrum. An important tool in the calculation is the cyclotomic trace of \textit{M. Bökstedt, W. C. Hsiang} and \textit{I. Madsen} [``The cyclotomic trace and algebraic K-theory of spaces'', Invent. Math. 111, No.~3, 465--539 (1993; Zbl 0804.55004)].