Nilpotenz und Auflösbarkeit topologischer Gruppen (Q790960)

Let A, B be subsets of a topological group G. The closed subgroup of G generated by the set \(\{aba^{-1}b^{-1}| a\in A,b\in B\}\) is denoted by [A,B]. We define the commutator series \(\{D^ kG\}\) and descending central series \(\{C^ kG\}\) of G as follows: \(D^ 0G=C^ 1G=G,D^ kG=[D^{k-1},D^{k-1}G], C^{k+1}G=[G,C^ kG]\) for \(k=1,2,...\). The author calls G topologically nilpotent (resp. solvable) iff for every neighborhood U of e (the unit element) in G there is \(k>0\) such that \(C^ kG\subset U\) (resp. \(D^ kG\subset U)\). A topological group G is called topologically reductive (resp. semi-simple) iff every connected, topologically solvable normal subgroup of G is contained in the center of G (resp. reduces to \(\{\) \(e\})\). A maximal connected topologically solvable normal subgroup of G, if any, is called the radical of G. Now, let \(G=\lim G_{\alpha}\) be a connected LP-group and \(L=\lim L_{\alpha}\) its Lie algebra in the sense of \textit{R. K. Lashof} [Pac. J. Math. 7, 1145-1162 (1957; Zbl 0081.022)]. The author gives (mutually) equivalent conditions for G to be topologically nilpotent, solvable, semi-simple, or reductive in terms of \(G_{\alpha}\), L or \(L_{\alpha}\). The author finally constructs a topological group G without radical.