The Coxeter Group [5,3,3,3] Has No Torsion-Free Subgroup of Index 14400: A Computer-Assisted Result Related to Problem 4.124 of the K3 List

Problem 4.124 of the K3 problem list asks whether a hyperbolic integer homology 4-sphere exists, whether an arithmetic one exists, and more generally whether there is a closed hyperbolic 4-manifold with Euler characteristic 2. By a theorem of Belolipetsky, an orientable arithmetic example would be the quotient of H^4 by a torsion-free subgroup of index 28800 of the Coxeter group W = [5,3,3,3] that lies in the rotation subgroup of W. We prove, with computer assistance, that W has no torsion-free subgroup of index 14400. This answers a question that goes back to a remark of Davis from 1985, who thought such a subgroup quite possible. Equivalently: no facet pairing of a single compact regular hyperbolic 120-cell with dihedral angle 2π/3 gives a closed hyperbolic 4-manifold; every torsion-free subgroup of finite index of W has index 14400·m with m ≥ 2 (Conder and Maclachlan found one with m = 8); and no closed manifold that covers the orbifold H^4/W has Euler characteristic 1. The proof reduces the question, with complete proofs, to the enumeration of certain involutions of the 14400 flags of the 120-cell. It then rests on two exhaustive enumerations, made by two separately written programs with different symmetry reductions and different pruning rules; both find nothing. No complete gluing was reached in either enumeration, so the result depends on every pruning rule being a necessary condition. We prove this for each of the 13 rules of the first program and each of the 14 rules of the second, and we report controls, mutation tests and partial re-enumerations with a third program. The theorem decides none of the three questions of Problem 4.124. It excludes only those subgroups of index 28800 that correspond to double covers of one-cell manifolds; the case of index 28800 itself is out of reach of our programs, and the problem stays open. This is an unrefereed note. Unrefereed preprint released for independent mathematical scrutiny. Publication on Zenodo does not constitute peer review. AI-assisted tools supported research, computation, proof development, and manuscript preparation. The author remains responsible for all claims and the final text. Corpus identifier: KP-4.124 (K3 problem list, Problem 4.124).

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-08
DOI
https://doi.org/10.5281/zenodo.23233096
Primary Topic
Geometric and Algebraic Topology
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

The Coxeter Group [5,3,3,3] Has No Torsion-Free Subgroup of Index 14400: A Computer-Assisted Result Related to Problem 4.124 of the K3 List

Alper Ferudun
Zenodo (CERN European Organization for Nuclear Research)
Geometric and Algebraic Topology
preprint

The Coxeter Group [5,3,3,3] Has No Torsion-Free Subgroup of Index 14400: A Computer-Assisted Result Related to Problem 4.124 of the K3 List

Alper Ferudun
preprint en

Abstract

Problem 4.124 of the K3 problem list asks whether a hyperbolic integer homology 4-sphere exists, whether an arithmetic one exists, and more generally whether there is a closed hyperbolic 4-manifold with Euler characteristic 2. By a theorem of Belolipetsky, an orientable arithmetic example would be the quotient of H^4 by a torsion-free subgroup of index 28800 of the Coxeter group W = [5,3,3,3] that lies in the rotation subgroup of W. We prove, with computer assistance, that W has no torsion-free subgroup of index 14400. This answers a question that goes back to a remark of Davis from 1985, who thought such a subgroup quite possible. Equivalently: no facet pairing of a single compact regular hyperbolic 120-cell with dihedral angle 2π/3 gives a closed hyperbolic 4-manifold; every torsion-free subgroup of finite index of W has index 14400·m with m ≥ 2 (Conder and Maclachlan found one with m = 8); and no closed manifold that covers the orbifold H^4/W has Euler characteristic 1. The proof reduces the question, with complete proofs, to the enumeration of certain involutions of the 14400 flags of the 120-cell. It then rests on two exhaustive enumerations, made by two separately written programs with different symmetry reductions and different pruning rules; both find nothing. No complete gluing was reached in either enumeration, so the result depends on every pruning rule being a necessary condition. We prove this for each of the 13 rules of the first program and each of the 14 rules of the second, and we report controls, mutation tests and partial re-enumerations with a third program. The theorem decides none of the three questions of Problem 4.124. It excludes only those subgroups of index 28800 that correspond to double covers of one-cell manifolds; the case of index 28800 itself is out of reach of our programs, and the problem stays open. This is an unrefereed note. Unrefereed preprint released for independent mathematical scrutiny. Publication on Zenodo does not constitute peer review. AI-assisted tools supported research, computation, proof development, and manuscript preparation. The author remains responsible for all claims and the final text. Corpus identifier: KP-4.124 (K3 problem list, Problem 4.124).

Zenodo (CERN European Organization for Nuclear Research)
Geometric and Algebraic Topology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.