Empty Horizon Black Holes: A Self-Limiting Mechanism for Cosmic Stability and the Universe's Decomposition Infrastructure

Empty Horizon Black Holes: A Self-Limiting Mechanism for Cosmic Stability and the Universe's Decomposition Infrastructure presents a complete theoretical framework for a new class of black hole formed not from gravitational collapse of matter, but from near-critical Coleman–De Luccia vacuum bubble nucleation events. When a CDL bubble nucleates at or above the critical radius derived from the Israel junction conditions, the self-gravitating bubble wall forms an apparent horizon before the post-nucleation field configuration decoheres — producing an Empty Horizon Black Hole whose interior is false vacuum rather than matter. The paper derives the semiclassical formation threshold from first principles, establishes the ghost-state criterion governing structural viability, and proves the Hawking One-Third Theorem: the fractional mass loss per containment cycle is exactly one-third, independent of remnant mass, a clean result following from the M³ scaling of both the cycle duration and the Hawking evaporation timescale. A gain function analysis identifies an unstable fixed point at the critical mass Mcrit, separating a damped regime (Regime 2, eventual metabolization) from an amplified regime (Regime 3, unbounded growth) — with Standard Model parameters placing Mcrit ≈ Mₑsc, making every astrophysical EHBH automatically a Regime 3 decomposer. The framework derives specific gravitational wave signatures for each regime, including a distinctive anti-chirp ladder and a Regime 0 truncated burst from failed formation events, both potentially accessible to LISA. In the far future of the universe, Regime 3 EHBHs seeded by evaporating gravitational collapse black holes near the escape mass serve as the universe's decomposition infrastructure — systematically processing false vacuum energy on timescales of 10⁷⁰ to 10¹⁰⁰ years, with a single EHBH per causal patch sufficient for the mechanism to operate. The paper includes a complete notation table, executive summary, worked numerical example, open questions with tractability ratings, a nine-prediction falsifiability summary, and a fully documented Python code package archived alongside the paper.