The α formula, piece by pieceEvery factor, traced back to the chapter that derived it

The coherence lattice, the CLR, the diamond, the vortex, the BKT wall, the PLM Lemma, the living-versus-static choice — twelve chapters of physics now fit into a single line of algebra. This chapter displays that line at full scale, with every symbol clickable. Each click opens a detail panel showing the lattice feature that produces it, the chapter where it is derived, and the role it plays in the formula. Nothing here is tuned; nothing is an input. Every piece is a measurable property of the electron as a phase vortex on the diamond lattice.

First — what is α?

Before looking at the formula, it helps to know what the number on the left-hand side actually is. In standard physics, the fine structure constant is the dimensionless strength of electromagnetism — a single number, roughly 1/137, that sits at the heart of four measurements. Any two of them, done with different apparatus, give the same value to twelve significant figures:

1. Coulomb force between electrons

The repulsive force between two electrons separated by distance r is F = α ℏc / r². Factor out the universal constants ℏ and c and what remains is α — the fundamental strength of the electromagnetic push.

2. Bohr radius of hydrogen

An electron's orbit around a proton has a characteristic size a₀ = ℏ / (m_ecα) ≈ 0.529 Å. Larger α means a tighter orbit. Atom size is a direct measurement of α.

3. Fine-structure splitting

Spectral lines in hydrogen and heavier atoms split into closely spaced doublets, separated by an amount proportional to α². Sommerfeld discovered this in 1916 and named α after it. Still the most accurate non-g-factor measurement.

4. Anomalous magnetic moment

The electron's magnetic moment is g/2 times a natural unit. Dirac predicted g = 2 exactly; experiment finds g/2 − 1 = α/(2π) + (higher orders). Agreement with theory: 12 matching digits — the most precise prediction in all of physics.

Notice what all four have in common. Each measurement probes a different physical thing — the force, the orbit, the spectrum, the spin. Yet each returns the same dimensionless number. That is what makes α a fundamental constant: it isn't a property of any one atom or any one experiment. It is the single number that sets the strength of electromagnetism, universally. Whatever underlying structure reality has, that structure must output this one number.

For a hundred years nobody has derived it from first principles. Eddington tried in 1929 and was ridiculed. Dirac famously called it "one of the greatest unsolved problems in physics". Feynman wrote: "It's been a mystery ever since it was discovered more than fifty years ago... we can measure it very accurately, but we don't understand it." The coherence lattice is, to our knowledge, the first framework that reproduces the measured value to 29 ppm with zero fit parameters.

The test of physics vs numerologyNumerologists have derived 1/137 from combinations of integers and π for a century. What makes the coherence lattice different is that every piece of the formula has an independent physical referent on the lattice — you can point at it, measure it, vary it, and watch the prediction change. If you remove the diamond lattice, you cannot have an exponent 4. If you remove BKT topology, the coupling 2/π doesn't appear. If you replace the CLR with fixed-K dynamics, the Debye–Waller factor collapses. The formula is not an arithmetic coincidence — it is the combined output of seven independent pieces of lattice physics, each measurable, each falsifiable.

The formula

α = R₀(2/π)⁴ × (π/4)^{1/√e + α/(2π)}

Click any coloured symbol above to see where it comes from, what it means, and the value it takes at the BKT critical point.

The single equation packages five independent physical inputs, each derived from the lattice dynamics and none of them fit to data. Click any coloured symbol above to see the lattice feature that produces it and to jump to the chapter where it is derived.

What α is, on the lattice

Each of the four measurements of α — the Coulomb force, the Bohr radius, the fine-structure splitting, the anomalous magnetic moment — probes a different physical quantity. On the coherence lattice, all four are faces of one dynamical quantity: the amplitude for a phase fluctuation in the K-field to propagate from the vacuum to the electron vortex and back.

Coulomb force ⇄ bond-coherence transmission

Two vortices separated by distance r feel each other through phase fluctuations of the bonds between them. The efficiency of that transmission is the product of per-bond correlations R₀^z (the vertex factor), running out to r via the BKT power law. That is α.

Bohr radius ⇄ vortex extent in lattice units

The electron vortex has a finite spatial scale set by the balance of phase-gradient energy and the K-field's healing length. Measured in units of the Compton wavelength, that scale is 1/α — a direct consequence of the vertex factor and the BKT marginality condition (Chapter 11).

Spectroscopic splitting ⇄ vortex self-interaction

Two energy levels of the same vortex differ by its self-interaction through the K-field — a second-order process giving α². Sommerfeld's 1916 constant is the lattice's 4-bond star-graph correlation, squared.

Anomalous g-factor ⇄ K-field vacuum response

For frozen K, the electron has g = 2 exactly (Dirac's result, Chapter 10). The anomaly g/2 − 1 = α/(2π) is the leading correction from the K-field's self-optimising response to the vortex — a virtual K-fluctuation absorbed and re-emitted. The tiny Schwinger term in the exponent of the formula is this same virtual loop.

The electron's charge, by contrast, is a different kind of quantity: an integer. It comes from the topological winding of the phase around the vortex core (Chapter 9), and it is quantised because winding numbers are quantised. Charge and α are not the same thing: charge is 1, 2, 3, … (times e); α is 1/137.036… always. But they share a home on the lattice: the electron is a phase vortex, the charge is the winding of its phase, and α is how that vortex couples to everything else through the K-field. Remove the lattice and both quantities lose their meaning.

So why this specific formula? Every factor in the expression above corresponds to one piece of the vortex's interaction with its lattice environment:

R₀(2/π) — the correlation a single bond transmits at the BKT-critical coupling. This is the von Mises single-bond statistic from Chapter 2.
The exponent 4 — each atom in the lattice has four bonds, so the vortex core has four independent information channels to the outside. This is the coordination number of diamond from Chapter 8.
The base π/4 — the fraction of an RG coarse-graining step captured by a single lattice hop. This is Chapter 11's variance-ratio identity.
The exponent 1/√e — the coherence that survives a single RG shell, evaluated at the CLR attractor where K is frozen. This is Chapter 12's living-vs-static choice.
The correction α/(2π) — the vertex correction from a virtual K-fluctuation loop, which is exactly the Schwinger term. The lattice produces QED as an emergent theory.

The formula's self-consistency — the α appearing inside its own exponent — is not a mathematical curiosity. It is the statement that the vacuum the electron lives in is made out of the same stuff as the electron itself, and both adjust to each other. Iterate, and a fixed point appears. That fixed point is the number:

Self-consistent iteration of the formula, starting from the bare UV guess α₀ = V = R₀(2/π)⁴ and reinserting the Schwinger correction α/(2π) at each step. The horizontal axis is a 1/α number line; the orange dot is the current iterate; the green dashed line is the CODATA measured value 1/α = 137.036. The first iteration undershoots (1/α = 118.3, purple); each subsequent step corrects toward the fixed point. Three steps reach 137.032 — within 29 ppm of the measured value. The BKT formula has no free parameters; its only observable is α itself, and it produces one self-consistent answer.

Each iteration refines the value of α that is plugged back into the Schwinger correction. The correction is tiny — α/(2π) is about 0.00116, roughly a part per thousand of the matching exponent — but its sign and magnitude determine whether the fixed point exists and where it lies. The fact that the iteration converges at all is non-trivial: it means the self-consistent QED vacuum-polarisation structure is compatible with the BKT power law at this specific coupling. Change any of the five factors and the fixed point moves.

To see which factors matter most, nudge them:

Each slider perturbs one factor of the formula by ±5% around its derived value. The resulting 1/α is displayed at the right and plotted as a coloured tick on the bottom ruler (CODATA 137.036 marked in green). R₀(2/π) dominates the sensitivity: a ±5% change sweeps 1/α through roughly 55 integer units, more than 25x the sensitivity of the Debye–Waller exponent. This is why the derivation effort is concentrated on the vertex factor — it is where the ppm precision of the BKT prediction lives.

The hierarchy is stark: any error in R₀ is amplified by the fourth power (the coordination number z = 4 from Chapter 8). Errors in the base factor are amplified linearly by the exponent n ≈ 0.61. Errors in 1/√e barely matter. The numerical value of α is governed, above all, by a single number: the von Mises order parameter of a single bond at coupling 2/π. Everything else is correction.

Zero free parameters, twelve chapters deep

The ledgerEvery ingredient of the BKT formula has a chapter-level provenance. R₀ is the von Mises order parameter on a single bond (Chapter 2). The coupling 2/π is the BKT marginal value forced by vortex topology (Chapter 11). The exponent 4 is the coordination number of the diamond lattice (Chapter 8). The base π/4 is the lattice-to-RG variance ratio (Chapter 5). The matching exponent 1/√e is the Debye–Waller factor at the CLR attractor (Chapter 12). The correction α/(2π) is standard Schwinger QED. None of them is a fit, a scale, or a unit convention. The only thing you have to believe to get α out of this framework is the inequality I ≥ 0 — the Coherence Theorem from Chapter 4.

Closing the last 3%

The value 1/α = 137.032 produced by the BKT formula is within 29 parts per million of the CODATA measurement, but not exact. The remaining 3%-of-a-percent comes from QED vacuum polarisation below the lattice crossover scale — standard physics that runs α_BKT from the matching scale down to Q = 0 (the Thomson limit). Chapter 15 walks through that linked-cluster expansion, which closes the gap to 1.5 parts per billion. The BKT formula is the hard part; the LCE is the polish.

Where this leaves us

Before we take the polish, one last surprise. The formula has a hidden dimensional dial — the coordination number z = d + 1 (for a tetrahedral coordination in d dimensions). Sliding that dial from d = 2 to d = 5 gives 1/α values from 36 to 1290. Only d = 3 produces the physical value. That is the next chapter — and it is the kind of demonstration that stops a reader in their tracks.