An algebraic model for rational \(\operatorname{SO}(3)\)-spectra (Q2409818)

For \(G\) a compact Lie group, \(G\)-equivariant cohomology theories are a fundamental tool in the study of spaces with \(G\)-action. Examples include equivariant \(K\)-theory, equivariant cobordism and the Borel construction. Just as in the non-equivariant setting, \(G\)-equivariant cohomology theories are represented by \(G\)-spectra. In turn, rational \(G\)-spectra represent those equivariant cohomology theories which take values in rational vector spaces. These rationalised categories are much simpler than the full categories, but contain much of the interesting equivariant behaviour. A long standing project, started by Greenlees, seeks to classify rational \(G\)-spectra (and hence rational \(G\)-equivariant cohomology theories) in terms of simple algebraic models. The classification is known for finite groups, for SO\((2)\) and products thereof, and O\((2)\). This paper considers the case of \(G=\text{SO}(3)\), the Lie group of rotations of \(\mathbb{R}^3\). The additional complexity over the previous known cases arises from the fact that the maximal torus is no longer normal. In this case the maximal torus is SO\((2)\), which has normaliser O\((2)\). The author provides a very neat solution to this problem, one that can be generalised to other groups and will be an important step towards giving a classification for an arbitrary group \(G\). The method is to show that the forgetful functor and coinduction adjunction can be made into a Quillen equivalence between the `toral part' of rational SO\((3)\)-spectra and a cellularisation (a right Bousfield localisation) of rational O\((2)\)-spectra. The `toral part' here refers to those spectra whose geometric isotropy is concentrated over subconjugates of SO\((2)\). The author then uses the classification of rational SO\((2)\)-spectra to give an algebraic model for the toral part of rational SO\((3)\)-spectra. The main result of the paper is the classification of rational SO\((3)\)-equivariant spectra in terms of an explicit algebraic model. The algebraic model is given in the paper and consists of three parts. The toral part, as described above; the dihedral part, which is similar to the dihedral part of the algebraic model for rational O\((2)\)-spectra; and the exceptional part, which behaves like the case of a finite group. The exceptional part refers to those spectra with geometric isotropy concentrated over the exceptional closed subgroups of SO\((3)\): \(D_4\), \(\Sigma_4\), \(A_4\), \(A_5\) and SO\((3)\) itself.