Q.C. Zhang Twistor Configuration Geometry
← All papers twistor 0.24% (m_μ) · 0.54% (m_τ)

Charged-Lepton Mass Predictions from Golden-Ratio Scaling

The logarithms of the charged-lepton Yukawa couplings satisfy golden-ratio scaling; combined with the electron-Yukawa closed form, this predicts m_μ and m_τ from π, φ, and v alone.

Published
DOI 10.5281/zenodo.19981197
Key relation
ln(1/y_μ) / ln(1/y_τ) ≈ φ

Abstract

Starting from the approximate formula ye(11/(2π))e4πy_e \approx (1 - 1/(2\pi))\,e^{-4\pi} for the electron Yukawa coupling [Zhang 2026e], we observe that the logarithms of the charged lepton Yukawa couplings satisfy golden-ratio scaling:

ln(1/yμ)ln(1/yτ)    φ,ln(1/ye)ln(1/yμ)    φ+1π2,\frac{\ln(1/y_\mu)}{\ln(1/y_\tau)} \;\approx\; \varphi, \qquad \frac{\ln(1/y_e)}{\ln(1/y_\mu)} \;\approx\; \varphi + \frac{1}{\pi^2},

where φ=(1+5)/2\varphi = (1+\sqrt{5})/2 is the golden ratio. Combined with the electron formula, these two rules predict all three charged lepton masses from π\pi, φ\varphi, and the Higgs vacuum expectation value vv: mμ=105.4m_\mu = 105.4 MeV (measured: 105.66 MeV, 0.24%) and mτ=1786.5m_\tau = 1786.5 MeV (measured: 1777.0 MeV, 0.54%). An independent check confirms: mτ/mμ=eπ1/πm_\tau/m_\mu = e^{\pi - 1/\pi} to 0.08%. The golden ratio connects to the Fibonacci chain structure of the mass sector in the super-flag framework on CP34\mathbb{CP}^{3|4} [Zhang 2026h]. The scaling does not extend to quarks: down-type quarks scale as ~2 rather than φ\varphi, and up-type quarks are disrupted by the top quark (yt1y_t \approx 1). We report the lepton predictions honestly and the quark non-extension equally honestly.

The empirical relation reported here is also reviewed in the companion DAEDALUS review [Zhang 2026 review], where it is classified as a Cabibbo-scenario regularity; in the broader Twistor Configuration Geometry (TCG) framework [Zhang 2026 TCG], the Perron eigenvalue φ\varphi of the Fibonacci transfer matrix appears as a structural pattern, and the precise map to the lepton mass operator (resolving the sign-inversion mμ/mτ<1m_\mu/m_\tau < 1 versus φ>1\varphi > 1) remains an open task.

DOI

https://doi.org/10.5281/zenodo.19981197