A lower bound for coherences on the Brown-Peterson spectrum (Q2492031)

The notion of \(E_\infty\) ring spectrum arose in the 1970's and was developed in work of J.~P.~May and his collaborators. It codifies the highly coherent multiplicative structure on many naturally occurring ring spectra such as those associated with \(K\)-theory and cobordism theories, cf. \textit{J. P. May} [\(E_\infty\) ring spaces and \(E_\infty\) ring spectra. With contributions by Frank Quinn, Nigel Ray, and Jorgen Tornehave. Lecture Notes in Mathematics. 577. (Berlin-Heidelberg-New York): Springer-Verlag. (1977; Zbl 0345.55007)]. In the past decade, obstruction theories for the existence of \(E_\infty\) structures on ring spectra were developed by \textit{A. Robinson} and \textit{S. Whitehouse} [Math. Proc. Camb. Philos. Soc. 132, No.2, 197-234 (2002; Zbl 0997.18004)], \textit{A. Robinson} [Structured ring spectra. Cambridge: Cambridge University Press. London Mathematical Society Lecture Note Series 315, 133-149 (2004; Zbl 1082.55005)] based on \(\Gamma\)-cohomology, which has many relations with André-Quillen cohomology of commutative algebras. A closely related theory was described by \textit{P. G. Goerss} and \textit{M. J. Hopkins} [Structured ring spectra. Cambridge: Cambridge University Press. London Mathematical Society Lecture Note Series 315, 151-200 (2004; Zbl 1086.55006)]. These theories were applied to produce such structures on Lubin-Tate spectra which lack precise geometric models. Applications to \(K\)-theory were described by \textit{A. Baker} and \textit{B. Richter} [Comment. Math. Helv. 80, No.4, 691-723 (2005; Zbl 1094.55010)]. The present paper mainly focuses on the important example of the Brown-Peterson spectrum \(BP\) for a prime \(p\). This is known to be a commutative ring spectrum (and even an \(A_\infty\) ring spectrum), however to date it has not been shown to admit an \(E_\infty\) structure. In contrast, \(MU\) is well-known to be an \(E_\infty\) ring spectrum even though \textit{P. Hu, I. Kriz} and \textit{J. P. May} [Homology Homotopy Appl. 3, No.2, 341-354, electronic only (2001; Zbl 0987.55009)] showed the natural map \(BP\longrightarrow MU_{(p)}\) cannot be an \(E_\infty\) map. The approach taken uses a certain filtration of the \(E_\infty\) operad to define for each \(n\) an \textit{\(n\)-stage structure} on a ring spectrum. There is an obstruction theory for passing from \((n-1)\) to \(n\)-stage structures, with obstructions lying in suitable \(\Gamma\)-cohomology groups. In the case of \(BP\), a careful analysis of the gradings involved shows that the possible obstructions vanish to allow at least a \((2p^2+2p-2)\)-stage structure. An interesting consequence is that the mod~\(p\) homology of \(BP\) supports a collection of low degree Dyer-Lashof operations and these are used to give a proof that \(BP\) cannot be realised as the Thom spectrum associated with an \(n\)-fold loop map to \(BSF\), where \(n=4\) if \(p=2\) and \(n=2p+2\) if \(p\) is odd. Although there is a stronger result of this type due to Priddy, the present approach seems elegantly conceptual rather than calculational. Further examples of periodic spectra where information about \(n\)-stage structures can be obtained are also mentioned.