OpenFrontiers
← Back to the index
OP-0459OpenAnalysis

The classification problem for amenable C-algebras · 1994-ana-174

Functional Analysis · ICM 1994: The classification problem for amenable C-algebras

Source assessment · not reviewedAs of 2026-08-18Source & evidence ↓

Problem statement

Prove that every nonzero, separable, amenable (equivalently, nuclear), unital, simple C∗C^*-algebra AA such that K0(A)=K1(A)=0K_0(A)=K_1(A)=0 and AA has no tracial states is isomorphic to the Cuntz algebra O2\mathcal O_2, the universal unital C∗C^*-algebra generated by two isometries s1,s2s_1,s_2 satisfying si∗sj=δij1s_i^*s_j=\delta_{ij}1 and s1s1∗+s2s2∗=1s_1s_1^*+s_2s_2^*=1.

Lean 4

Not yet attached

No Lean statement or proof is attached yet. Have a formalization? Share its source to add it to this problem.

Remaining questions in this record

Remaining questions recorded by the source

These are retained within the parent record and are not counted as additional index entries.

Question 1

Prove or disprove that every unital, separable, nuclear, simple, purely infinite C∗C^*-algebra AA such that K0(A)=K1(A)=0K_0(A)=K_1(A)=0 is isomorphic to the Cuntz algebra O2\mathcal O_2, the universal unital C∗C^*-algebra generated by two isometries s1,s2s_1,s_2 satisfying si∗sj=δij1s_i^*s_j=\delta_{ij}1 and s1s1∗+s2s2∗=1s_1s_1^*+s_2s_2^*=1.

Source reference: Christopher Schafhauser, Aaron Tikuisis, and Stuart White, “Nuclear C∗C^*-Algebras: 99 Problems,” preprint, 2025, revised 2026, Section 3 (the equivalent zero-KK-theory Kirchberg-algebra formulation following Problem II), arXiv:2506.10902, https://arxiv.org/abs/2506.10902.

Question 2

Determine whether there exists a separable, unital, simple, nuclear C∗C^*-algebra of real rank zero that contains both an infinite projection and a nonzero finite projection.

Source reference: Mikael Rørdam, “A Simple C∗C^*-Algebra with a Finite and an Infinite Projection,” Acta Mathematica 191 (2003), no. 1, 109-142, Question 7.6(ii), DOI: 10.1007/BF02392697, arXiv:math/0204339, https://arxiv.org/abs/math/0204339.

Progress & source assessment

Source-reported assessment from the ICM catalogue snapshot. This is not an independent review by Frontier Science.

The problem is unresolved. No proof or counterexample is known under exactly the stated hypotheses.

The strongest complete positive result requires extra hypotheses. If AA is also purely infinite and satisfies the Universal Coefficient Theorem (UCT), the Kirchberg-Phillips classification theorem identifies unital Kirchberg algebras by their graded KK-groups and the class of the unit. Thus K0(A)=K1(A)=0K_0(A)=K_1(A)=0 implies A≅O2A\cong\mathcal O_2. Mikael Rørdam further proved that a separable nuclear Z\mathcal Z-stable algebra is O∞\mathcal O_\infty-stable exactly when it is traceless. Consequently, the assertion holds when both UCT and Z\mathcal Z-stability are added.

Even the narrower case in which AA is assumed purely infinite remains open without UCT. Rørdam and Størmer, Corollary 8.4.6(ii), show that the assertion that every unital Kirchberg algebra with K∗(A)=0K_*(A)=0 is isomorphic to O2\mathcal O_2 is equivalent to the UCT problem for nuclear C∗C^*-algebras. Schafhauser, Tikuisis, and White explicitly record this equivalence and the continuing open status in their current survey. Hence the original statement, which does not assume UCT, contains a major unresolved special case.

There is also no general route from the original no-trace hypothesis to pure infiniteness. Rørdam constructed a simple, separable, nuclear UCT algebra containing both a nonzero finite projection and an infinite projection; it has no tracial state but is not purely infinite. This disproves the general implication “traceless implies purely infinite,” although that construction does not establish a counterexample having both KK-groups zero. Thus it does not resolve the exact problem.

Key references:

  • George A. Elliott, “The Classification Problem for Amenable C∗C^*-Algebras,” Proceedings of the International Congress of Mathematicians, Zürich 1994, Birkhäuser, 1995, pp. 922-932, DOI: 10.1007/978-3-0348-9078-6_85. Section 6 states the exact proposed characterization of O2\mathcal O_2.
  • N. Christopher Phillips, “A Classification Theorem for Nuclear Purely Infinite Simple C∗C^*-Algebras,” Documenta Mathematica 5 (2000), 49-114, arXiv:funct-an/9506010. It proves the UCT purely infinite case via the Kirchberg-Phillips classification theorem.
  • Mikael Rørdam and Erling Størmer, Classification of Nuclear C∗C^*-Algebras. Entropy in Operator Algebras, Encyclopaedia of Mathematical Sciences 126, Springer, 2002, DOI: 10.1007/978-3-662-04825-2. Corollary 8.4.6(ii) gives the equivalence between the UCT problem and uniqueness of O2\mathcal O_2 among unital Kirchberg algebras with zero KK-theory.
  • Mikael Rørdam, “A Simple C∗C^*-Algebra with a Finite and an Infinite Projection,” Acta Mathematica 191 (2003), 109-142, DOI: 10.1007/BF02392697, arXiv:math/0204339. Theorem 6.10 and Corollary 7.1 give a separable nuclear UCT example with both finite and infinite projections, showing that absence of traces alone does not force pure infiniteness.
  • Mikael Rørdam, “The Stable and the Real Rank of Z\mathcal Z-Absorbing C∗C^*-Algebras,” International Journal of Mathematics 15 (2004), 1065-1084, arXiv:math/0408020. Theorem 5.2 proves that separable nuclear Z\mathcal Z-stable traceless algebras are O∞\mathcal O_\infty-stable, giving the principal regularity-based partial result.
  • Christopher Schafhauser, Aaron Tikuisis, and Stuart White, “Nuclear C∗C^*-Algebras: 99 Problems,” preprint, 2025 (revised 2026), arXiv:2506.10902. Section 3 records that the zero-KK-theory Kirchberg special case is equivalent to the still-open UCT problem.
References
  1. [1]ICM Conjectures · 1994-ana-174 link
  2. [2]ICM 1994: The classification problem for amenable C-algebras link

Recorded progress

Add yours →

Nothing recorded yet. A pointer to a paper the entry is missing counts.

Machine attempts

0 recorded

Attempts run against the statement below and are recorded in full, including the failures. The point is to stop the next model — and the next person — from re-walking a road that is already known to end. Each attempt costs 25 credits.

No machine attempts recorded.

Sign in to run an attempt. New accounts start with enough credits for ten.

Discussion

0 notes

Notes are for clarifying the statement, flagging a missing reference, or recording why an approach fails. Claimed solutions belong in a submission, where they can be reviewed.

No notes yet.

Sign in to add a note.