The group of \(K_{1}\)-zero-cycles on the second generalized Severi-Brauer variety of an algebra of index 4 (Q2400335)

Let \(X\) be an algebraic variety over an infinite perfect field \(F\) of characteristic different from \(2\). It is known (see [\textit{D. Quillen}, Lect. Notes Math. 341, 85--147 (1973; Zbl 0292.18004)] and [\textit{M. Rost}, Doc. Math., J. DMV 1, 319--393 (1996; Zbl 0864.14002)] that there is a complex \(\dots \to \sqcup_{x \in X _{(p+1)}} K_{p+q+1}(F(x)) \to\) \(\sqcup_{x \in X _{(p)}} K_{p+q}(F(x)) \to \) \(\sqcup_{x \in X _{(p-1)}} K_{p+q-1}(F(x)) \to \dots \), where \(X_{(i)}\) is the set of points of \(X\) of dimension \(i\), \(K_i\) are the Quillen \(K\)-groups (see [\textit{V. Srinivas}, Algebraic \(K\)-theory. Paperback reprint of the 1996 2nd edition. Boston, MA: Birkhäuser (2008; Zbl 1125.19300)]), and the differentials are given component-wise by residue homomorphisms. Denote by \(A_p(X, K_q)\) the pth homology group of this complex, i.e., homology at the middle term above. It is known (due to [\textit{A. S. Merkurjev} and \textit{A. A. Suslin}, Nova J. Algebra Geom. 1, No. 3, 297--315 (1992; Zbl 0870.19001)]) that if \(X\) is the Severi-Brauer variety associated to a central simple algebra \(A\), then the group \(A_ 0(X, K_1)\) of \(K_ 1\)-zero-cycles is given by \(K_1(A)\). The paper under review provides a similar result for cycles on generalized Severi-Brauer varieties, i.e. twisted forms of Grassmannians (see [\textit{A. Blanchet}, Commun. Algebra 19, No. 1, 97--118 (1991; Zbl 0717.16014)]). It shows that the group \(A _0(X, K_1)\), where \(X\) is the second generalized Severi-Brauer variety of an algebra \(A\) of index \(4\) (and arbitrary degree), is given by elements of \(K _(A)\) together with a square root of their reduced norm. More precisely, \(A _0(X,K_1)\) is isomorphic to the group \(\{(x,\alpha ) \in K_ 1(A) \times F^{\ast } | \text{Nrd}_{A}(x) = \alpha^2\}\). To prove their main theorem, the authors use results of Krashen concerning exceptional isomorphisms, and translate the problem to the computation of cycles on involution varieties. Work of Chernousov and Merkurjev then gives them means of describing such cycles in terms of Clifford and spin groups and corresponding \(R\)-equivalence classes. An analysis of these groups completes the proof.