Residually finite-dimensional \(C^*\)-algebras and subquotients of the CAR algebra (Q5953008)

From MaRDI portal
Revision as of 23:47, 4 March 2024 by Import240304020342 (talk | contribs) (Set profile property.)
scientific article; zbMATH DE number 1690767
Language Label Description Also known as
English
Residually finite-dimensional \(C^*\)-algebras and subquotients of the CAR algebra
scientific article; zbMATH DE number 1690767

    Statements

    Residually finite-dimensional \(C^*\)-algebras and subquotients of the CAR algebra (English)
    0 references
    0 references
    1 October 2002
    0 references
    A separable \(C^*\)-algebra is called residually finite dimensional (RFD) if it has a separating sequence of finite dimensional representations. The author first proves that for such a separable nuclearly embeddable RFD \(C^*\)-algebra \(A\), the cone \(CA =\) \(C_0[0,1)\otimes A\) of \(A\) and the suspension \(SA=\) \(C_0(0,1)\otimes A\) of \(A\) are embeddable in the CAR algebra \(B\) (= the UHF algebra of type \(2^{\infty}\)) (Theorem 2.6): as for the CAR algebra, see e.g. \textit{E. Kirchberg} [J. Funct. Anal. 129, No. 1, 35-63 (1995; Zbl 0912.46059)]. Next it is proved that for a separable unital nuclearly embeddable \(C^*\)-algebra \(A\), there exists a semisplit essential extension \(0 \rightarrow\) \(I \rightarrow\) \(\widetilde{CF} \rightarrow\) \(A \rightarrow 0\) with a unital separable nuclearly embeddable RED \(C^*\)-algebra \(F\) (Lemma 3.1), where \(\widetilde{A}\) denotes the \(C^*\)-algebra obtained by adding a unit to \(A\). Combining the above two basic results, the author finally gives a new short proof of Kirchberg's theorem [cf. E. Kirchberg, op. cit.], which asserts that a separable unital \(C^*\)-algebra \(A\) is nuclearly embeddable if and only if there is a semisplit extension \(0 \rightarrow\) \(J \rightarrow\) \(E \rightarrow\) \(A \rightarrow 0\) such that \(E\) is a unital \(C^*\)-subalgebra of the CAR algebra \(B\) and \(J\) is an AF-algebra stably isomorphic to \(B\) (theorem 1.1). The remarkable feature in this paper consists in the fact that the new proof does not rely on the lifting theorem of \textit{E. G. Effros} and \textit{U. Haagerup} [Duke Math. J. 52, 103-128 (1985; Zbl 0613.46047)]. The proofs of the basic results (Theorem 2.6 and (resp. Lemma 3.1)) are due to the author's previous results: Proposition 2.3 in [J. Funct. Anal. 167, No. 1, 69-78 (1999; Zbl 0936.46043)] and resp. Proposition 3.3 in [the author, ibid. 151, 213-233 (1997; Zbl 0992.46042)].
    0 references
    0 references
    nuclearly embeddable \(C^*\)-algebra
    0 references
    CAR algebra
    0 references
    residually finite-dimensional
    0 references
    semispit essential extension
    0 references
    Kirchberg's theorem
    0 references

    Identifiers