Quantitative Supercritical Bounds for Disconnection in Bernoulli Site Percolation
Zhongyang Li
TL;DR
The paper addresses the problem of obtaining quantitative disconnection tails in supercritical Bernoulli site percolation on general infinite graphs. It introduces the $(p,\varepsilon,c)$-packing number $\mathrm{PK}_{p,\varepsilon,c}(S)$ together with the local functional $\varphi_p^v(S)$ to connect finite-volume data to the global percolation threshold, and proves an explicit exponential-type bound on $\mathbb{P}_{p}(S\nleftrightarrow\infty)$ that scales with the packing size. The method combines a finite-volume connectivity analysis (FCA) with an independence amplification across well-separated packing vertices, yielding a bound of the form $\mathbb{P}_p(S\nleftrightarrow\infty) \le \frac{(1-c)\varepsilon}{c}+(1+\varepsilon)(1-c)^{\mathrm{PK}_{p,\varepsilon,c}(S)}$ and its infimum representation. By situating the result within existing effective supercritical bounds (capacity-based, isoperimetric, and conjectural isoperimetric tails), the work provides a unified, quantitative framework for disconnection tails driven by packing structure.
Abstract
For any infinite, connected, locally finite graph $G=(V,E)$, any parameter $p>p^{\mathrm{site}}_{c}(G)$, and any (finite or infinite) set of vertices $S\subset V$, we derive explicit exponential-type upper bounds on the disconnection probability $\mathbb{P}_{p}(S\nleftrightarrow\infty)$. The estimates are expressed in terms of a packing profile of $S$, encoded by a $(p,\varepsilon,c)$--packing number, which counts how many well-separated vertices in $S$ exhibit controlled local-to-global connectivity. The proof combines a local functional characterization of $p^{\mathrm{site}}_{c}$ from \cite{ZL24,ZL26} with a packing construction and an amplification-by-independence argument, in the direction of Problem~1.6 in \cite{DC20}.
