Thermodynamics of the SGCC Bounce: Bekenstein–Hawking Entropy Conservation, the Horizon Budget, and the Refutation of Heat Death [v2 — Erratum: torsion amplitude calibration]

ERRATUM (v2, 2026-07-04) — Correction to the torsion amplitude |T0| calibration (Appendix D). A review of the torsion amplitude calibration revealed three corrections affecting the quantitative interpretation of the bounce, without altering the Riemann-Cartan framework or the central heat-death refutation result. (1) BOUNCE REGIME. The ad-hoc value |T0| = 1e-4 H0^2 places the bounce at a_bounce ~ 0.04-0.07 (z ~ 14-22), i.e. AFTER CMB recombination (z=1100) and Big Bang Nucleosynthesis (BBN). This regime is observationally excluded. The binding bound is not the Solar System limit (|T0| < 3e-4) but BBN preservation: |T0| eps_pv^2 < Omega_r a_BBN^2 ~ 5e-24, nineteen orders of magnitude stricter. (2) SINGLE PARAMETER. The amplitude |T0| and the parity parameter eps_pv enter the Friedmann equation only through the product P = |T0| eps_pv^2; they are not separately observable. Bounce phenomenology depends on P alone, with viable window P in [3.6e-68, 5.1e-24] (pre-BBN, sub-Planckian; further bounded above by Poplawski's beta < 1/929 against eternal inflation). (3) TORSION SOURCE. The first-principles calibration of |T0| from the LoTSS DR2 SMBH spin density (Appendix D) used the coherent vector sum of angular momentum, <S>. The standard Einstein-Cartan-Sciama-Kibble spin-fluid formalism uses the QUADRATIC term <S^2> (Weyssenhoff; Poplawski 2016, ApJ 832, 96), independent of orientation coherence. More fundamentally, the physical source of the bounce torsion is the primordial fermionic fluid at densities > 1e45 kg/m^3, not the SMBH population, which forms ~18 orders of magnitude in scale factor AFTER the bounce. The value |T0|_SMBH ~ 8.6e-41 H0^2 must be interpreted as a scale analogy of the spin-torsion coupling in the accessible regime, NOT as the causal source of the bounce. WHAT DOES NOT CHANGE. The Riemann-Cartan framework, the stiff scaling (a0/a)^6, the result that any bounce-compatible torsion term is undetectable in the CMB by construction (scale decoupling), and consistency with the null Q22xCMB cross-correlation limits (Papers VIII/IX) remain valid and are in fact strengthened: the CMB null is trivial by construction, not a constraint on eps_pv. The final sentence of the original abstract below ("...providing a theoretical justification for the null CMB result... 36 orders below the Solar System bound") is superseded by points (1) and (3) above: |T0|_SMBH is a scale analogy, not a first-principles source calibration, and the CMB null is trivial rather than a constraint. A complete treatment of the viability window and the observational anchoring of the SMBH population within the natural cosmological selection framework (Smolin 1992) + torsional bounce (Poplawski) is presented in separate work (in preparation). ───────────────────────────────────────────────── ORIGINAL ABSTRACT (v1, unchanged): We present a complete thermodynamic analysis of the Self-Generated Cyclic Cosmology (SGCC) bounce within the Riemann–Cartan framework. Using the Bekenstein–Hawking entropy of the supermassive black hole (SMBH) population as the primary thermodynamic observable, we demonstrate that the classical heat death of the universe is not a physical fate within the SGCC framework, but a limit that is never reached because the torsional bounce interrupts the cosmic timeline 81 orders of magnitude before Hawking evaporation completes. We compute six independent thermodynamic quantities. (i) The Hawking evaporation timescale tau_H ~ 3.6×10^91 yr for a median SMBH mass of 1.2×10^8 M_sun, compared to a typical bounce time t_bounce ~ 7 Gyr, giving tau_H/t_bounce ~ 2.6×10^81. (ii) Area conservation through the Kruskal throat: DeltaS_BH/S_BH < 10^-13, confirming the horizon area as a topological invariant across the bounce. (iii) The horizon budget: the entropy density s = S_BH/V_horizon is 10^2–10^3 times higher at the bounce than today, demonstrating that the universe is far from thermodynamic equilibrium at the bounce. (iv) Pre-bounce accretion: DeltaM/M < 10^-15 — mass and entropy are conserved to numerical precision. (v) The adiabatic cycle invariant J = S_BH/S_rad = 3.617×10^-16, constant across the bounce, establishes thermodynamic self-consistency of the SGCC cycle. (vi) The von Neumann entropy of the torsion field S_vN ~ 10^42 J K^-1 quantifies the information encoded in the spin-memory mechanism of Paper B.