The center of the generic \(G\)-crossed product (Q2630883)

In the study of the Brauer group \(\mathrm{Br}({\mathbb F})\) of a field \(\mathbb F\), an important tool is the generic division algebra \(D_n\) of degree \(n\) over \(\mathbb F\). It gives also information about the groups \(\mathrm{Br}({\mathbb L})\) for field extensions \({\mathbb L}/{\mathbb F}\) and has the remarkable property that any other central simple algebra of degree \(n\) of a field extension of \(\mathbb F\) is a specialization of \(D_n\). The celebrated Merkur'ev-Suslin theorem [\textit{A. S. Merkur'ev} and \textit{A. A. Suslin}, Math. USSR, Izv. 21, 307--340 (1983); translation from Izv. Akad. Nauk SSSR, Ser. Mat. 46, No. 5, 1011--1046 (1982; Zbl 0525.18008)] states that if \(\mathbb F\) contains enough roots of unity, then every central simple algebra is equivalent to a tensor product of cyclic algebras. If one could show that the center of \(D_n\) is a rational extension of \(\mathbb F\), then it will follow that any central simple algebra of degree \(n\) over a field extension of \(\mathbb F\) is equivalent to a product of cyclic algebras, thus proving Merkur'ev-Suslin's theorem for such fields. Motivated by the significance of the center \(Z_n := Z(D_n)\) of \(D_n\), an important problem is whether the field extension \(Z_n/{\mathbb F}\) is rational (i.e., purely transcendental), or has some weaker rationality property, namely it is stably rational, retract rational or just unirational. Many results in this direction have been obtained starting with the rationality of \(Z_2/{\mathbb F}\) in the paper ``On the involution of two matrices of the second order'' by Sylvester in 1883 (reprinted in [\textit{J. J. Sylvester}, The Collected Mathematical Papers of James Joseph Sylvester. Volume IV (1882--1894). Cambridge: University Press (1912; JFM 43.0026.01)]). Since any central simple algebra is Brauer equivalent to a \(G\)-crossed product for some finite group \(G\), it is natural to consider generic \(G\)-crossed products. In the paper under review, the author gives a construction a generic \(G\)-crossed product over \(\mathbb F\) using generic graded matrices (\(\text{char}\,{\mathbb F}=0\)). The center \({\mathbb F}(G)\) of this generic \(G\)-crossed product is the invariant field of a suitable \(G\)-action on a field of rational functions in several indeterminates. The main goal of this paper is to study the extensions \({\mathbb F}(G)/{\mathbb F}\) given that \(\mathbb F\) contains enough roots of unity and to determine how close they are to be purely transcendental. In particular, \({\mathbb F}(G)/{\mathbb F}\) is a retract rational extension for \(G=C_n \rtimes C_m\) with coprime \(n\) and \(m\). The extension \({\mathbb F}(G)/{\mathbb F}\) is stably rational for \(G=C_2 \times C_{2n}\) when \(n\) is odd and for \(G=\langle \sigma,\tau| \sigma^n=\tau^{2m}=e,\tau\sigma\tau=\sigma^{-1}\rangle\) with coprime \(n\) and \(2m\). The author also proves that if \(H\) and \(K\) are groups of coprime orders, then \({\mathbb F}(H\times K)\) is stably rationally equivalent to the fraction field of \({\mathbb F}(H)\otimes{\mathbb F}(K)\).