\(v_ n\)-elements in ring spectra and applications to bordism theory (Q1362082): Difference between revisions

The paper consists of two parts. The first part deals with a generalization of the \(v_n\)-element, defined by Hopkins and Smith, in a finite \((p\)-local) ring spectrum. Suppose \(n \geq 0\) and let \(R\) be a \((p\)-local) ring spectrum. According to the author's definition, then an element \(v\in \pi_*(R)\) is called a \(v_n\)-element if \(K(n)_* (v^j) =v^k_n\) for some \(j\), \(k\), and \(K(i)_*(v)\) is nilpotent for \(i>n\). Here \(K(n)\) denote the Morava \(K\)-theories. This definition is motivated by the fact that \(v_n\in BP_*\) is not a \(v_n\)-element in the sense of Hopkins and Smith. It is shown that this generalized \(v_n\)-element in a finite ring spectrum is exactly an ordinary \(v_n\)-element when \(n>0\) and so this definition is reasonable. One of the main results of this part is a theorem on a sufficient condition for the existence of such an element in a ring spectrum. This is given by using the \(v_n\)-elements whose existence is already known in a finite ring spectrum. In particular, this condition works well on the case where \(p\) is odd and \(n=1\). For example, the author proves that, if the homotopy class \(\alpha_t\in \pi_{2t(p-1)-1} S^0\) maps to zero under the unit map to \(R\), then there is a \(v_1\)-element in \(R\). Analyzing the telescope conjecture for \(R\), the author also shows that, if a \(v_1\)-element \(v\) in \(R\) is given, then the localization of \(R\) at \(K(1)\) can be obtained by inverting \(v\) and completing at \(p\), that is, \(L_{K(1)} R=(v^{-1} R)_p\). Let \(MO\langle k\rangle\) and \(MU\langle k\rangle\) denote the Thom spectra arising from the \((k-1)\)-connected Postnikov covers of \(BO\) and \(BU\) respectively. The second part is devoted to applications to these Thom spectra. Using the \(\alpha\)-family which belongs to the image of the \(J\)-homomorphism, the author constructs \(v_1\)-elements \(v\) in \(MO\langle k\rangle\) and \(MU\langle k\rangle\). From the fact described above we then see that each element \(v\) yields \(L_{K(1)}R=(v^{-1} R)_p\) for \(R=MO\langle k\rangle\), \(MU\langle k\rangle\). Further the author considers \(L_{K (1)}R\) in another aspect and gives the following extended result of a theorem of \textit{M. A. Hopkins} and the author [Math. Z. 210, No. 2, 181-196 (1992; Zbl 0770.55008)]. (Here we take \(MO\langle k\rangle.)\) If the torsion in \(\pi_*M O\langle k\rangle\) is bounded, then the natural map \(MO\langle k\rangle_* (X) \otimes_{MO\langle k\rangle_*} KO_*\to KO_*(X)\) is an isomorphism. The proof uses two kinds of the above descriptions of \(L_{K(1)} MO\langle k\rangle\) and the result of Hopkins and the author.