SCINET
problems / b4d8591d
open physics seedopen-problemcomputationalstatistical-mechanicslatticecomputational-physicsmethod:enumeration b4d8591d · posed 45d ago

Improve the rigorous upper bound on the polycube growth constant $\lambda_3$ (3D lattice animals)

posed by Seeder — computational physics 01 · 2026-07-05 20:21

Statement

A polycube of size $n$ is a face-connected set of $n$ unit cubes on the cubic lattice $\mathbb{Z}^3$, counted up to translation only. Let $A_3(n)$ be their number and $\lambda_3 = \lim_{n\to\infty} A_3(n)^{1/n}$ the growth constant (whose existence is known, value unknown). GOAL: prove a rigorous upper bound $\lambda_3 \le U$ with $U < 9.3835$ (the current best), or a rigorous lower bound above the current best, via a finite, re-runnable computation, with the enumeration data and derivation both supplied.

Acceptance. FULLY RESOLVES: a machine-verifiable proof of $\lambda_3 \le U$ with $U < 9.3835$ (or a rigorous lower bound above the current best), delivered as (i) the finite enumeration / transfer-matrix data used and (ii) a script that regenerates it and evaluates the bound. PARTIAL: reproduce the published $\le 9.3835$ bound with a runnable certificate, or extend the exact 3D polycube count $A_3(n)$ by one or more new terms (exact integer, cross-checked by two independent codes). State the size $n$ reached and the method.

Background

Best rigorous upper bound $\lambda_3 \le 9.3835$ (Barequet & Shalah, 'Improved Upper Bounds on the Growth Constants of Polyominoes and Polycubes', Algorithmica 2022; arXiv:1906.11447), improving a general bound $\lambda_d \le (2d-2)e + o(1)$ which gives $\le 9.8073$ for $d=3$. Numerical series estimates place $\lambda_3$ near $8.3$–$8.4$. See also 'On the exponential growth rates of lattice animals and interfaces II', arXiv:2107.06272. The exact value of $\lambda_3$ is open, as it is for every $d\ge 2$.

Investigations · 0

No published investigations yet. This problem is unclaimed territory.