The twisted commutator subgroups and questions of Fel'shtyn and Troitsky (Q2659564)

There is a notion of a group \(G\) being conjugacy separable and for an automorphism \(\phi\) of \(G\) there is a similar notion called \(\phi\)-conjugacy separable where the conjugacy classes are replaced by the Reidemeister classes of \(\varphi\). The paper under review first shows that ``the group of the rationals \(\mathbb{Q}\) is not residually finite but it is \(\varphi\)-conjugacy separable for any automorphism \(\varphi\ne Id\).'' Then they introduce the notion of \textit{twisted commutator subgroup} with respect to an automorphism \(\phi\) which is defined as the subgroup generated by the elements of the form \(xy\phi(x)^{-1}y^{-1}\) and denoted by \(G_{\phi}\). They show that \(G_{\phi}\) is normal and they prove some results about the Reidemeister number of \(\phi\) and the property of \(G\) be \(\phi\)-conjugacy separable in terms of this subgroup \(G_{\phi}\) and the quotient \(G/G_{\phi}\). Just to illustrate, one of the results is: Proposition 3.8: Let \(\phi:G\to G\) be an automorphism of \(G\) with \(R(\phi)<\infty\). If \(G_{\phi}\) is twisted conjugacy separable, such that \(G_{\phi}\) and \(G/G_{\phi}\) are finitely generated, then \(G\) is \(\phi\)-conjugacy separable.