Round 2 verification audit for C061

Model: gpt-5

Claim: Spontaneous emission from |e> with optical-qubit lifetime T_1 limits an MS gate of length t to a Bell-state error ~= 0.5 t/T_1, and the amplified error gives a lower bound on the protocol steady-state infidelity of ~= 3 t/T_1.

Source alignment: source/supp_content.tex:117-117 (Supp. Mat. S3)

Prior official verdict: partial with failure_reason None.

Executable evidence: sympy_check.py. Sandbox rerun logs: sympy_check.log.

Independent audit: I scanned the copied script for imports/shared helper dependencies and reran it through the sandbox. The code is self-contained in this attempt directory and targets the claim strategy: Derive the MS-gate spontaneous-emission error ~0.5 t/T1 and propagate it through the protocol amplification factor to recover the ~3 t/T1 lower bound.. I checked the relevant family model rather than relying only on exit status; the rerun is treated as one reproducibility input.

Decision: Round 2 verdict is partial with failure_reason None and limitations ['paper_text_only_reimplementation']. Notes: Paper: MS-gate Bell error ~0.5 t/T1; protocol steady-state infidelity lower bound ~3 t/T1. Derivation: expected decays over a gate = _avg * t/T1 with _avg ~ 0.5 (final |ee> has 1 excited ion, population rises 0->1), reproducing 0.5 t/T1 as an order-of-magnitude estimate (linear-area numeric 0.36, same order). the self-contained model reproduces the amplification factor exactly: steady-state singlet error = 3.18 p for per-cycle correlated bit-flip p (= main-text 3.2p). The ~3 t/T1 bound = amplification(3.18) x per-cycle decay(~t/T1); consistent in form but the factor 3 vs gate 0.5 requires the amplification model plus an O(1) per-cycle decay assumption, not a single clean identity. Both stated approximately. -> partial.