Torsion completeness of Sylow \(p\)-groups in semisimple group rings. (Q1780301): Difference between revisions

Let \(KG\) be the group algebra of an Abelian \(p\)-group \(G\) over a field \(K\) of characteristic different from \(p\). Suppose \(K\) is a field of the first kind with respect to \(p\) and the spectrum \(s_p(K)\) of \(K\) with respect to \(p\) is \(\mathbb{N}\) or \(\mathbb{N}\cup\{0\}\), where \(\mathbb{N}\) is the set of the natural numbers. Let \(S(KG)\) be the \(p\)-component of the group of the normalized units of \(KG\). In his main result (Theorem 1) the author tries to prove that \(S(KG)\) is torsion complete if and only if \(G\) is bounded. But this article is incorrect since it contains a series of lapses and ungrounded conclusions in the proof of Theorem 1, namely: (i) From the representation of the element \(x_b\) (p. 897, line 9 from above) we see that \(KB'\) contains only two minimal orthogonal idempotents \(e_1\) and \(e_2\). This is possible only under some special conditions on the group \(B'\) and the field \(K\) but such conditions are not met in the proof and in the formulation of the theorem. This is the most essential mistake in the proof. (ii) The author states absolutely ungroundedly, that the equality on page 896, line 10 from above, is equivalent to the equality on page 896, line 11 from above. (iii) The author incorrectly writes (p. 896, lines 18-19 from above): ``From our initial discussion in the introduction we with no harm of generality assume that \(\text{height}(g_1)\geq m\), \dots'', and this fact remains unchecked, since it is not considered in the introduction. (iv) In fact the author does not prove case 3 in Theorem 1, when the group \(B\) is uncountable and he only writes ``The assertion follows by standard transfinite induction \dots''. But this is inadmissible at the indicated defects of the article. Other lapses and mistakes in the article are the following: (v) The decomposition of \(S(KF)\) (p. 894, line 19 from above) is incorrect. (vi) In the proof of the Proposition (Structure) the author states incorrectly: ``Clearly, \(G\) and \(G'\) have the same infinite Ulm-Kaplansky invariants, \dots'' (page 895, lines 16-15 from below), since \(G'\) is infinite and \(G\) can be finite. (vii) The author writes that the results can be proved ``without the restriction on the spectrum on the first kind field \(K\)'' (page 898, line 23 from below) which, because of the note in (i), is untrue. In this way Theorem 1 and therefore Corollaries 1-3 and the Global Theorem remain unproved. So almost all results of the article are incorrect. When \(K\) is a field of the second kind with respect to \(p\) the author proves Theorem 2, but it is a trivial corollary of a result of the reviewer, cited as [1] in the References [PLISKA, Stud. Math. Bulg. 8, 34-46 (1986; Zbl 0662.16008)]. The author writes down vaguely and impedes the reading. He does not give the main concept of the spectrum \(s_p(K)\) of \(K\) with respect to \(p\), that is introduced by the reviewer in the article [1] in Russian in a Bulgarian journal along with a constant of a field \(K\) with respect to \(p\).