The direct factor problem for modular group algebras of isolated direct sums of torsion-complete Abelian groups. (Q1434298)

Let \(RG\) be the group ring of an Abelian group \(G\) over a commutative ring \(R\) of prime characteristic \(p\) and let \(S(RG)\) be the \(p\)-component of the group of the normalized units of \(RG\). The author tries to prove that if \(G_p\) is an isolated direct sum of torsion-complete groups then \(G_p\) is a direct factor of \(S(RG)\) but he does not manage to prove it. This paper is unsuccessful. The introduced concept ``isolated direct sum of subgroups'' has absolutely no sense and is superfluous. Namely it is easy to see that if \(A_p\) is an isolated direct sum of subgroups then it is a direct sum of subgroups and, conversely, if \(A_p\) is a direct sum of subgroups \(A_i\) then it is an isolated direct sum of subgroups \(C_i=A_i\). In this concept the groups \(C_i\) are absolutely superfluous. It is strange that the author does not understand this since in the ``Concluding Discussion'' of the paper he raises the following ``interesting problem'': ``\dots does it follow that each direct sum of torsion-complete groups can be embedded in an isolated direct sum of torsion-complete groups?'' The proofs of the main results are incorrect, unclear and ungrounded. The author beforehand tries to prove a Group Proposition and two lemmas. The Group Proposition is simply superfluous since its formulation is the following: ``If \dots \(A_p\) is a direct sum of torsion-complete groups, then \(A_p\) is isomorphic to an isolated direct sum of torsion-complete groups.'' The first lemma is not true. We give two counterexamples. Let \(G=(a)\) be a cyclic group of order \(p^2\), where \(p\) is a prime. Let \(C=(a^p)\) and (i) \(M=\{a,a^{p-1}\}\). Then \(C\cap M=\emptyset\) and \(I(RC;C\cap M)\) has no sense. (ii) Now, let us set \(M=\{1,a,a^{p-1}\}\). Then \(C\cap M=\{1\}\) and the right hand side of the formula of the lemma is \(1+I(RC;C\cap M)=1\) but the left hand side contains the element \(a^p=1+a^{p-1}(a-1)+a^{p-1}-1\neq 1\). This lemma is used essentially in the proof of the Major Proposition (Decomposition). The second lemma is also incorrect. It would be corrected if we add the conditions \(1\in M\) and \(1\in N\). The Major Proposition (Decomposition) is also untrue. We can give the following counterexample. Let \(C_p\neq 1\) and \(a\) is an arbitrary element from \(C_p\). Suppose \(g\in G\setminus(C\times C')\). Therefore \(g-1\in I=I(RG;C' \times[(G\setminus(C\times C'))\cup\{1\}])\). Since \(g^{-1}\notin C\times C'\) and \(a\in C\times C'\) then \(g^{-1}a\notin C\times C'\), i.e. \(g^{-1 }a\in G\setminus(C\times C')\). Hence \(g^{-1}a-1\in I\). Then \(a-1=g^{-1}a(g-1)+g^{-1}a-1\in I\). Since \(a-1\) is a nilpotent element then \(a-1\in I_p=I_p(RG;C' \times[(G\setminus(C\times C'))\cup\{1\}])\). Then \(a\in 1+I_p\). But, obviously, \(a\in S(RC)\). Therefore \(S(RC)\cap(1+I_p)\) contains the group \(C_p\), i.e. the indicated product in the first formula of the Major Proposition (Decomposition) is not direct. This formula, which is not true, is used in the proof of the second formula of the proposition. Besides, in the proof of this proposition it is untrue to assert that \(G_p\) is a direct sum of groups \((D_\beta)_p\). In the formulation of the Major Proposition (Decomposition) is used the sign \(\times\) for sets which has a sense in the case only for groups. The incorrectly proved Major Proposition (Decomposition) is used in the proof of the Major Theorem (Direct Factor). In this way the main results established in the paper are much weaker and remain unproved in the generality stated by the author.