English

A new source of purely finite matricial fields

Group Theory 2026-04-14 v5 Functional Analysis Operator Algebras Probability Spectral Theory

Abstract

A countable group GG is said to be \emph{matricial field} (MF) if it admits a strongly converging sequence of approximate homomorphisms into matrices; i.e, the norms of polynomials converge to those in the left regular representation. GG is \emph{purely MF} (PMF) if these maps are actual homomorphisms, and GG is further \emph{purely finite field} (PFF) if the image of each homomorphism is finite. By developing a new operator algebraic approach to these problems, we are able to prove the following result bringing several new examples into the fold. Suppose GG is a MF (resp., PMF, PFF) group and H<GH<G is separable (i.e., H=iNHiH=\cap_{i\in \mathbb{N}}H_i where Hi<GH_i<G are finite index subgroups) and KK is a residually finite MF (resp., PMF, PFF) group. If either GG or KK is exact, then the amalgamated free product GH(H×K)G*_{H}(H\times K) is MF (resp., PMF, PFF). Our work has several applications, we list some below: 1. The Brown--Douglas--Fillmore semigroups of many new examples of reduced group CC^*-algebras are shown to be not groups. 2. Arbitrary group doubles GHGG*_HG of MF (resp., PMF, PFF) over separable subgroups HH are MF (resp., PMF, PFF). Moreover, GHG*H is PFF whenever G,HG,H are PFF, and either GG or HH is exact. 3. Arbitrary graph products of residually finite exact MF (resp., PMF, PFF) groups are MF (resp., PMF, PFF), yielding a significant generalization of the breakthrough work of M. Magee and J. Thomas. 4. The open problem of proving PFF for fundamental groups of closed hyperbolic 3-manifolds is resolved. This has geometric significance in the theory of minimal surfaces via A. Song's approach.

Keywords

Cite

@article{arxiv.2603.24502,
  title  = {A new source of purely finite matricial fields},
  author = {David Gao and Srivatsav Kunnawalkam Elayavalli and Aareyan Manzoor and Gregory Patchell},
  journal= {arXiv preprint arXiv:2603.24502},
  year   = {2026}
}

Comments

15 pages. For Vidhya Ranganathan. v2: Fixed typo in Corollary 1.4. v3: improved readability; added section 1.3: optimality of results, explicitness/concreteness of matrix models, further bootstrapping, insights into proof; added more details and explanation in Lemma 2.3. v4. minor changes and expanded on geometric significance, thanks to A. Song. v5. minor changes around references