Products of C*-algebras that do not embed into the Calkin algebra
Damian Głodkowski, Piotr Koszmider · 2024
Read Products of C*-algebras that do not embed into the Calkin algebra by Damian Głodkowski, Piotr Koszmider (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-11191.txt.
Introduction
Damian Glodkowski and Piotr Koszmider study whether a big algebra can fit inside a famous quotient algebra, in a 2024 paper. In one consistent world of set theory, they show it cannot. The Calkin algebra bundles all well-behaved operators on a standard infinite-dimensional space, modulo the compact ones. Whether a given algebra embeds in it is a subtle test of size and structure.
Working inside the Cohen model of set theory, the authors prove a negative result. The endless product of one abelian algebra with itself admits no such embedding, though the single algebra always does. The result joins a growing family of algebras that consistently refuse the Calkin home. It underlines how set-theoretic assumptions shape operator algebra theory.
Catalog record
| CREATOR | Damian Głodkowski, Piotr Koszmider |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-11191
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Products of C*-algebras that do not embed into the Calkin algebra Thanks: The first-named author was partially supported by the NCN (National Science Centre, Poland) research grant no. 2021/41/N/ST1/03682. Thanks: The second-named author was partially supported by the NCN (National Science Centre, Poland) research grant no. 2020/37/B/ST1/02613. Thanks: For the purpose of Open Access, the authors have applied a CC-BY public copyright license to any Author Accepted Manuscript (AAM) version arising from this submission. Damian Głodkowski, Piotr Koszmider Abstract . We consider the Calkin algebra 𝒬 ( ℓ 2 ) \mathcal{Q}(\ell_{2}) , i.e., the quotient of the algebra ℬ ( ℓ 2 ) \mathcal{B}(\ell_{2}) of all bounded linear operators on the separable Hilbert space ℓ 2 \ell_{2} divided by the ideal 𝒦 ( ℓ 2 ) \mathcal{K}(\ell_{2}) of all compact operators on ℓ 2 \ell_{2} . We show that in the Cohen model of set theory ZFC there is no embedding of the product ( c 0 ( 2 ω ) ) ℕ (c_{0}(2^{\omega}))^{\mathbb{N}} of infinitely many copies of the abelian C*-algebra c 0 ( 2 ω ) c_{0}(2^{\omega}) into 𝒬 ( ℓ 2 ) \mathcal{Q}(\ell_{2}) (while c 0 ( 2 ω ) c_{0}(2^{\omega}) always embeds into 𝒬 ( ℓ 2 ) \mathcal{Q}(\ell_{2}) ). This enlarges the collection of the known examples due to Vaccaro and to McKenney and Vignati of abelian algebras, asymptotic sequence algebras, reduced products and coronas of stabilizations which consistently do not embed into the Calkin algebra. As in the Cohen model the rigidity of quotient structures fails in general, our methods do not rely on these rigidity phenomena as is the case of most examples mentioned above. The results should be considered in the context of the result of Farah, Hirshberg and Vignati which says that consistently all C*-algebras of density up to 2 ω 2^{\omega} do embed into 𝒬 ( ℓ 2 ) \mathcal{Q}(\ell_{2}) . In particular, the algebra ( c 0 ( 2 ω ) ) ℕ (c_{0}(2^{\omega}))^{\mathbb{N}} consistently embeds into the Calkin algebra as well. 1. Introduction For a separable Hilbert space H we denote by \mathcal{B}(H) the C*-algebra of all bounded linear operators on H and by \mathcal{K}(H) the ideal of \mathcal{B}(H) consisting of compact operators. The quotient \mathcal{Q}(\ell_{2})=\mathcal{B}(\ell_{2})/\mathcal{K}(\ell_{2}) is called the Calkin algebra. The significance of the Calkin algebra for the understanding of the algebra \mathcal{B}(\ell_{2}) has been realized since its inception in [ 5 ] . This paper concerns the investigation of the class of all C*-algebras which can be embedded 1 1 1 Unless stated otherwise, by an embedding we mean an injective * -homomorphism of C*-algebras. into \mathcal{Q}(\ell_{2}) (necessarily of density up to 2^{\omega} as the density of \mathcal{Q}(\ell_{2}) is 2^{\omega} ). This class is denoted \mathbb{E} following [ 34 ] . It follows from the result of [ 30 ] which says that it is consistent that no Banach space of density 2^{\omega} contains isometric copies of all Banach spaces of density 2^{\omega} (“isometric” here may even be replaced by “isomorphic” by [ 3 ] ), that it is consistent that some abelian C*-algebra of density 2^{\omega} is not in \mathbb{E} (as every Banach space isometrically embeds into the C*-algebra of continuous functions on its dual ball with the weak ∗ topology). However, the witnesses of the nonuniversality in [ 30 ] or [ 3 ] are fundamentally nonconstructive. Concrete consistent examples of (abelian) C*-algebras of density 2^{\omega} not in \mathbb{E} have been produced at least since the doctoral thesis of Vignati [ 35 ] which was followed by other algebras or other consistent hypotheses in [ 25 ] , [ 32 ] , [ 34 ] . On the other hand, on the positive side, it is not difficult to see that \mathcal{B}(\ell_{2}) (obviously with many of its subalgebras) is in \mathbb{E} , and so all separable C*-algebras belong to the class \mathbb{E} (see e.g. [ 2 , Corollary II.6.4.10] ). Perhaps the simplest example of a C*-algebra that does not embed into \mathcal{B}(\ell_{2}) but does embed into \mathcal{Q}(\ell_{2}) is the abelian algebra c_{0}(2^{\omega}) . The existence of such an embedding follows from the existence of an uncountable almost disjoint (i.e., with finite pairwise intersections) family of infinite subsets of \mathbb{N} . Moreover, it was shown by Farah, Hirshberg and Vignati in [ 16 ] that all C*-algebras of density equal to the first uncountable cardinal \omega_{1} are in \mathbb{E} . In particular, under the continuum hypothesis CH , the Calkin algebra is universal in the class of C*-algebras of density 2^{\omega} . Also Martin’s axiom implies that all C*-algebras of density strictly less than 2^{\omega} belong to \mathbb{E} ( [ 17 ] ). The main purpose of this note is to provide concrete examples of C*-algebras that consistently do not embed into the Calkin algebra, i.e., are not in \mathbb{E} which are not covered by the results of [ 35 ] , [ 25 ] , [ 32 ] , [ 34 ] , are obtained by rather different methods, and exist in the absence of the combinatorial or forcing principles considered in [ 35 ] [ 25 ] , [ 34 ] and more interestingly, in the absence of the general rigidity of quotient structures: Theorem 1 . In the Cohen model 2 2 2 By the Cohen model we mean a model obtained from a model of CH by adding \omega_{2} Cohen reals. the abelian C*-algebra (c_{0}(2^{\omega}))^{\mathbb{N}} does not embed into the Calkin algebra \mathcal{Q}(\ell_{2}) . As a consequence the following types of C*-algebras do not embed into the Calkin algebra: • products \prod_{n\in\mathbb{N}}\mathcal{A}_{n} , • reduced product algebras \prod_{n\in\mathbb{N}}{\mathcal{A}_{n}}/\bigoplus_{\mathcal{I}}{\mathcal{A}_{n}}, • the corona algebras {\mathcal{Q}}({\mathcal{A}}\otimes{\mathcal{K}}) of the stabilization {\mathcal{A}}\otimes{\mathcal{K}} of \mathcal{A} , where each of the C*-algebras {\mathcal{A}},{\mathcal{A}}_{n} admits a pairwise orthogonal family of projections of cardinality 2^{\omega} and \mathcal{I} is any ideal of subsets of \mathbb{N} such that the Boolean algebra \wp(\mathbb{N})/\mathcal{I} is infinite. Proof. See Theorem 28 , Corollary 29 and Corollary 31 . ∎ Reduced products of C*-algebras have been considered at least in [ 35 ] , [ 25 ] , [ 15 ] , [ 14 ] , [ 20 ] , [ 21 ] , while the coronas of stabilizations were considered in the context of embeddings into the Calkin algebra in [ 25 ] , [ 34 ] , [ 18 ] . It is also interesting to compare the Calkin algebra with other corona algebras: Farah shows in [ 18 ] that \mathcal{Q}(\ell_{2}) is not isomorphic to the corona of the stabilization of the Cuntz algebra i.e. \mathcal{Q}(\mathcal{O}_{\infty}\otimes\mathcal{K}(\ell_{2})) , though these algebras are not distinguishable from the K-theoretical point of view. This research is followed in [ 19 ] . As mentioned in [ 18 ] the problem of the existence of such isomorphisms is important, since they may induce K-theory reversing automorphisms of \mathcal{Q}(\ell_{2}) (such automorphisms are not known to exist consistently). Let us compare our Theorem 1 to some of the results of the papers mentioned above. Concerning the products and direct sums one may make some conclusion from the results in [ 34 ] , where it is proved that the tensor products \mathcal{Q}(\ell_{2})\otimes\mathcal{A} do not embed into \mathcal{Q}(\ell_{2}) for any