\(I\)-adic towers in topology (Q813179): Difference between revisions

The celebrated Landweber exact functor theorem has prominently been applied in stable homotopy theory for many years which motivated to study analogues and related aspects in brave new algebra. Given a commutative \(S\)-algebra \(R\) like \(MU\), an \(R\)-algebra \(T\) and an ideal \(I \subset T_*\), generated by a regular sequence, one may ask for a topological realization of the tower \(T_* \to \dots \to T_*/I^{s+1} \to T_*/I^s \to \dots \to T_*/I\), which is called the algebraic \(I\)-adic tower under \(T_*\). \textit{A. Baker} and \textit{U. Würgler} already realized such a tower in the ordinary category of spectra in [Forum Math. 3, No.6, 543-560 (1991; Zbl 0751.55002)] in the case of \(R=MU\), the complex bordism spectrum, \(T=E(n)\), the Johnson-Wilson theory for a prime \(p\) and \(T_*/I=K(n)_*\), the coefficients of the corresponding \(n\)-th Morava \(K\)-theory. Several years later \textit{A. Baker} and \textit{A. Lazarev} presented a corresponding construction for a rather general class of \(S\)-algebras \(R\) in terms of brave new algebra in [Algebr. Geom. Topol. 1, 173--199 (2001; Zbl 0970.55006)]. In the paper under review the author presents a related but somewhat different construction of an \(I\)-adic tower under \(T\) in the context of brave new algebra, which has the advantage that his construction is a little more transparent, and that it provides a more conceptual understanding of the tower. The author proceeds as follows. First he considers the sequence \(\dots \to I^{s+1} \to I^{s} \to \dots \to T_*\) as a sequence in the derived category of \(T_*\)-modules. This way he obtains a sequence of degree \((-1)\) morphisms \(T_*/I \to I/I^2 \to I^2/I^3 \to \dots\) which (using terminology from [\textit{J. D. Christensen}, Adv. Math. 136, No.2, 284-339 (1998; Zbl 0928.55010)]) can be viewed as a relative injective resolution of \(T_*\) with respect to the injective class associated to \(T_*/I\). He then shows that the original tower over \(T_*\) can be reconstructed from the latter as its Adams resolution. Based on this observation he starts with realizing a corresponding sequence of fibrations \(T/I \to I/I^2 \to \dots\) in brave new algebra, where each of the fibrations \(I^{s-1}/I^{s} \to I^{s}/I^{s+1}\) is naturally classified by higher Bockstein operations. The associated \(I\)-adic tower over \(T\) is then defined to be the associated Adams resolution, while the desired \(I\)-adic tower under \(T\) is obtained from the latter by taking cofibers of the individual maps \(I^s \to T\) occurring in the \(I\)-adic tower over \(T\). The \(I\)-adic tower under \(T\) naturally gives rise to a conditionally convergent spectral sequence converging to \((holim_s T/I^s)^*(X)\) for a space \(X\), which in the special case of \(R=MU\), \(T=E(n)\) and \(T_*/I= K(n)_*\) specializes to the Bockstein spectral sequence of \textit{A. Baker} and \textit{U. Würgler} introduced in [op. cit.].