# Hydrogen Matrix Lab 0.7.0 — golden-ratio exactness audit

## Question

Where does the golden ratio genuinely occur in the current framework, and which apparent alignments are imposed choices or numerical coincidences?

The protocol was frozen in `GOLDEN_RATIO_PROTOCOL_0.7.json` before execution. It distinguishes exact symbolic derivation, construction-imposed use, finite numerical approximation, and measured evidence in another declared physical model.

## Exact definitions

The audit uses

\[
\varphi=\frac{1+\sqrt5}{2},\qquad
\varphi^2=\varphi+1,
\]

and the golden-turn fraction and golden angle

\[
\alpha=\frac1{\varphi^2}=\frac{3-\sqrt5}{2}=0.381966\ldots,
\qquad
\theta_g=2\pi\alpha=137.50776405\ldots{}^\circ.
\]

The production engine represents elements of \(\mathbb Q(\sqrt5)\) as exact rational pairs \(a+b\sqrt5\), so the polynomial identities are checked without floating-point inference.

## Exact manifestations

1. The Fibonacci companion matrix

   \[
   Q=\begin{pmatrix}1&1\\1&0\end{pmatrix}
   \]

   has characteristic polynomial \(x^2-x-1\), eigenvalues \(\varphi\) and \(-1/\varphi\), and

   \[
   Q^n=\begin{pmatrix}F_{n+1}&F_n\\F_n&F_{n-1}\end{pmatrix}.
   \]

   This is an intrinsic algebraic occurrence.

2. Standard icosahedral coordinates use permutations of \((0,\pm1,\pm\varphi)\). The computed ratio of the next vertex separation to the edge length is \(\varphi\). This is exact for the declared geometry, but the geometry itself was chosen.

   The familiar golden rectangle, regular pentagon, and golden spiral use the same constant rather than defining independent golden ratios. The pentagon diagonal-to-side ratio is \(2\cos(\pi/5)=\varphi\). A golden rectangle selects side ratio \(\varphi\) by definition. A logarithmic spiral is golden only after choosing \(b=2\log(\varphi)/\pi\) in \(r(\theta)=ae^{b\theta}\), so each quarter turn scales the radius by \(\varphi\).

3. The first two mass scales in the integrable `E8` spectrum obey \(m_2/m_1=2\cos(\pi/5)=\varphi\). Coldea et al. reported experimental neutron-scattering evidence consistent with this relation in a near-critical Ising-chain material. That is measured evidence for that system and theory, not for hydrogen or an extra spatial dimension.

## Exact hydrogen exclusion

For the ideal fixed-proton Coulomb spectrum,

\[
E_n=-\frac{1}{2n^2}\ {\rm Hartree}.
\]

Every \(E_n\) is rational in Hartree. Therefore every nonzero transition gap is rational, and a ratio of two such gaps is rational. Because \(\varphi\) is irrational,

\[
\boxed{\frac{E_b-E_a}{E_d-E_c}\ne\varphi}
\]

for every pair of nonzero ideal Coulomb bound-state gaps. This is an all-level written proof; it does not depend on the finite search.

The frozen finite computation enumerated all 780 transitions with \(1\le n<m\le40\) and 303,809 distinct gap-pair comparisons. The closest match was

\[
\frac{\Delta E_{19\to30}}{\Delta E_{23\to34}}
=\frac{7491169}{4629825}
=1.61802422337777\ldots,
\]

with absolute residual \(9.7653721\times10^{-6}\). It is explicitly retained as a near coincidence, not an exact law.

## Fixed-grid consequence

One angular step on a finite \(N\)-state grid is \(k/N\) turns and is rational. The exact golden turn \((3-\sqrt5)/2\) is irrational. Therefore an exact golden-angle step cannot live on a finite angular grid or have an exact finite recurrence period.

Fibonacci fractions provide controlled periodic approximations. The last frozen entry,

\[
\frac{987}{2584},
\]

misses the golden turn by \(6.69777\times10^{-8}\), or \(2.41120\times10^{-5}\) degrees per step. It is still a 2,584-tick rational surrogate, not the irrational rotation.

This matters for the Theory of Reality's fixed-grid/fixed-tick hypothesis: it must choose between an exactly finite angular state space and an exact golden rotation, or specify a larger algebraic/infinite-precision state representation. It cannot silently claim both.

## External measured manifestations and limits

- Douady and Couder produced Fibonacci/phyllotactic ordering in a laboratory experiment and numerical simulation, with convergence toward the golden mean through a self-organizing process.
- Shechtman et al. measured a metallic solid with icosahedral long-range orientational order and no ordinary translational lattice.
- Coldea et al. measured an excitation ratio consistent with the golden ratio predicted by emergent `E8` symmetry in a near-critical Ising chain.

These demonstrate that golden structures can arise in nature when a specific dynamical or symmetry mechanism selects them. They do not show that every spiral, atom, harmonic, or projection is golden.

Primary sources:

- S. Douady and Y. Couder, *Physical Review Letters* 68, 2098 (1992): https://doi.org/10.1103/PhysRevLett.68.2098
- D. Shechtman, I. Blech, D. Gratias, and J. W. Cahn, *Physical Review Letters* 53, 1951 (1984): https://doi.org/10.1103/PhysRevLett.53.1951
- R. Coldea et al., *Science* 327, 177 (2010): https://doi.org/10.1126/science.1180085

## Verification and evidence classes

- Written mathematics: exact \(\mathbb Q(\sqrt5)\) identities; Fibonacci eigenvalue; ideal-hydrogen rationality exclusion; finite-grid irrationality exclusion.
- Finite computation: 303,809 gap-pair comparisons, Fibonacci convergence table, icosahedral distances, and 144-point golden-angle prefix.
- Independent computation: a separate scalar implementation imports none of the production module and passes nine checks, including an independent exhaustive hydrogen scan.
- Measured evidence: only the cited external experiments; this release performs no new physical measurement.
- Hypothesis: a hidden-dimensional \(\varphi\) mechanism remains unspecified and unsupported by this audit.

The full package contains 51 unit tests after adding the nine release-0.7 tests. Earlier rotor rejections, spectroscopy corrections, probability-current results, and provenance remain unchanged.

## Next test

Do not add \(\varphi\) as a free visual parameter. First derive one dimensionless golden prediction from a declared Hamiltonian or projection operator that was not used to choose the model. Freeze the observable, uncertainty, and rejection threshold, then compare it against a non-golden baseline on fresh data. A decimal near-match after searching many ratios is not an admissible prediction.