Realizer Bounds and Repetition Constraints at the First Coefficient Crossing in the 3x + 1 Problem

A mix of human-written and AI-generated textHuman understanding: some partsmath.NT — Number Theorymath.CO — Combinatoricsmath.DS — Dynamical Systems

Contributed by Elias De Jesús

Version 2 / Nov 15, 2026 / CC BY 4.0

Abstract

Terras’s coefficient stopping time conjecture asks whether the shortcut Collatz orbit of every integer n ≥ 2 lies below n when its coefficient ρk = 3d/2k first falls below one. Here k counts shortcut steps and d counts their odd inputs. We prove that every eligible realizer descends when the first-crossing word has at most 75 maximal runs of odd inputs. Any counterexample requires at least 76 such runs, 10 439 860 591 steps and 6 586 818 670 odd inputs. The proof uses a lower bound on the seed from the length of its odd runs, rational approximations to log2 3, the verified range of Rozier and Terracol, and Rhin’s estimate for a linear form in logarithms. We also prove descent for a family with arbitrarily many odd runs. Two further restrictions narrow the remaining words. An initial square U U in a failing word satisfies |U | < 35 log2 k. For a counterexample of shortest crossing length, the reciprocal-sum estimate of Nikzad, Frinkle and Bagherzadeh gives a lower bound on ρk whose value begins 0.9441733669 . . .. The open inequality compares the least eligible starting value in each remaining parity class with an explicit threshold.

Provenance statement

This technical note was developed by Elias De Jesús, Independent Researcher, as part of an ongoing investigation of the Collatz problem, building on earlier work on least positive realizers, parity vectors, rational-shadow bounds, and symbolic repetition. DOI: https://doi.org/10.5281/zenodo.24231556 The research was motivated by Kazunobu Hikawa's November 2026 exposition of Terras's coefficient stopping time conjecture. It combines the author's existing realizer framework with published results of Terras, Rozier and Terracol, Rhin, and Nikzad, Frinkle and Bagherzadeh. Original contributions, mathematical deductions, and externally imported results are distinguished in the manuscript. Claude (Anthropic) and ChatGPT (OpenAI) were used under the author's direction for mathematical exploration, proof development and review, literature comparison, exact-arithmetic verification, and LaTeX manuscript preparation. Python scripts using integer and rational arithmetic were developed to reproduce numerical inequalities, verify finite counting results, and test representative Collatz trajectories. The verification code and supporting materials are provided in the companion reproducibility archive and associated GitHub repository: https://github.com/innerlightr-wq/terras-first-crossing The author assumes responsibility for the mathematical claims, interpretations, and any errors. The manuscript has not undergone independent peer review. No Lean formal verification of the results is claimed. The work establishes partial restrictions on potential counterexamples but does not prove Terras's conjecture or the Collatz conjecture.

References

  1. Justin Alick. Collatz mm-cycles: a L ean 4 formalisation, 2026. Declaration allPairs70 : AllPairs 70 76 in Collatz/MCycleTwoRiseHi2.lean lines 15--19, proved by decide +kernel , at commit d1895945 . https://github.com/justinalick/collatz-m-cycles-101-lean.link
  2. Shalom Eliahou. The 3x+13x+1 problem: new lower bounds on nontrivial cycle lengths. Discrete Mathematics , 118:45--56, 1993.
  3. Lynn E. Garner. On the C ollatz 3n+13n+1 algorithm. Proceedings of the American Mathematical Society , 82(1):19--22, 1981.
  4. Christian Hercher. There are no C ollatz-mm-cycles with m≤91m\le91. Journal of Integer Sequences , 26(3), 2023. Article 23.3.5. Corrigendum (Theorem 21), https://cs.uwaterloo.ca/journals/JIS/VOL26/Hercher/corrigendum.pdf; the conclusion m≥92m\ge92 is unaffected.link
  5. Kazunobu Hikawa. Research notes on Terras ' coefficient stopping time problem I : Previous results and open problems, 2026. English translation, November 2026; Japanese original, Zenodo, 13 November 2026, DOI 10.5281/zenodo.24187839.DOI
  6. Elias De Jesús. Beyond S turmian words in the 3x+13x+1 periodicity conjecture: A periodic-approximation and arithmetic-height bridge, 2026. Revision 5, 13 November 2026; Revision 4 deposited as Zenodo DOI 10.5281/zenodo.23070093.DOI
  7. Elias De Jesús. Exact realizer congruences and rational-shadow bounds for C ollatz words, 2026. Zenodo preprint, 31 October 2026. doi:10.5281/zenodo.23847289 (https://doi.org/10.5281/zenodo.23847289).DOI
  8. Elias De Jesús. Exact realizer floors for one- and two-defect valuation words in the accelerated 3x+13x+1 map, 2026. Zenodo preprint, 22 October 2026. doi:10.5281/zenodo.23579193 (https://doi.org/10.5281/zenodo.23579193).DOI
  9. Elias De Jesús. Independent verification of high-path reciprocal bounds for nonrepeating C ollatz trajectories, 2026. Unpublished manuscript, 3 November 2026.
  10. Jeffrey C. Lagarias. The 3x+13x+1 problem and its generalizations. American Mathematical Monthly , 92(1):3--23, 1985.
Show all 15 referencesShow fewer references
  1. Behzad Nikzad, Karl Frinkle, and Fatemeh Bagherzadeh. On bounds for reciprocal sums and residues in C ollatz orbits, and their consequences, 2026. Hexagon preprint 2611.00033v1, https://hexagonmath.org/2611.00033v1.hexagon
  2. Georges Rhin. Approximants de P adé et mesures effectives d'irrationalité. In Séminaire de Théorie des Nombres, Paris 1985--86 , volume 71 of Progress in Mathematics , pages 155--164. Birkhäuser Boston, 1987.
  3. Olivier Rozier and Claude Terracol. Paradoxical behavior in C ollatz sequences. Discrete Mathematics , 349:115167, 2026. arXiv:2502.00948v5.arXiv
  4. John Simons and Benne de Weger. Theoretical and computational bounds for mm-cycles of the 3n+13n+1 problem. Acta Arithmetica , 117(1):51--70, 2005. The bound was subsequently raised; Hercher hercher2023 reports m≥76m\ge76 for Simons--de Weger, and m≥83m\ge83 on substituting a modern verification range.
  5. Riho Terras. A stopping time problem on the positive integers. Acta Arithmetica , 30:241--252, 1976.

Version history

  1. v1 · 6 pagesSubmitted by Elias De JesúsInitial depositSupersededNov 14, 2026
  2. v2 · 12 pagesSubmitted by Elias De JesúsSubmitted as a new versionCurrentNov 15, 2026