Pauli Exclusion in the Quaternion Framework The Double Cover Gives the Sign; the Tensor Product Gives the Rule
The Pauli exclusion principle has exactly two ingredients, and this note assigns each to its rightful owner. There is a sign — the fact that a two-fermion state changes sign when the particles are exchanged — and there is a rule — that no two fermions may occupy the same state, which is the consequence of that sign. This series can claim the sign as genuinely quaternionic, and should claim nothing more. Fermions are half-windings of the quaternion sphere: the rotation group of space is doubly covered by the unit quaternions SU(2), in which a 4π turn is the identity and a 2π turn is −1, a fact measured directly in neutron interferometry in 1975. By the Finkelstein–Rubinstein argument, exchanging two identical such objects is equivalent to rotating one of them by 2π — and so multiplies the joint state by −1. That minus sign is the whole quaternionic dividend, and it is a real one. The rule is then standard quantum mechanics, and this note states it correctly: exclusion lives in the tensor product of single-particle Hilbert spaces, where the two-fermion state is the antisymmetrised (Slater) combination |α⟩∧|β⟩, whose squared norm 1 − |⟨α|β⟩|² vanishes if and only if the two single-particle states are the same ray, and is nonzero for every distinct pair. The quaternion earns a third, humbler role: it is an excellent label for a single electron's quantum numbers, packing energy, charge, spin, and a complex angular momentum onto its four axes — a legible identity card for |α⟩. What it does not do, and what an earlier draft of this series wrongly asked of it, is perform the antisymmetrisation: a binary product of two four-component labels — in particular the quaternion commutator ½(QP − PQ), which is merely the cross product of the imaginary parts and is blind to the energy axis — cannot reproduce the exclusion rule, because the rule depends on the inner product ⟨α|β⟩ of the full states, which no product of labels computes. That correction is made here once and then left behind; the construction below is built correctly from the start. The honest summary:
Abstract
1Two Ingredients, Two Owners
The exclusion principle forbids two identical fermions from sharing a quantum state, and from that single rule flow the periodic table, the stability of matter, and the degeneracy pressure of white dwarfs. It is usually stated as a postulate. Underneath the postulate are two separable facts. First, a sign: the joint wavefunction of two identical fermions is antisymmetric under their exchange, Ψ(1,2) = −Ψ(2,1). Second, a rule that follows from the sign: set the two particles in the same state and Ψ(1,1) = −Ψ(1,1) = 0 — the amplitude vanishes, so the configuration is forbidden. To 'explain' Pauli is to explain the sign; the rule is then automatic. This note keeps the two apart, because their origins are different: the sign is where the quaternion genuinely enters, and the rule is standard Hilbert-space bookkeeping.
2The Quaternion as a State Label
Give each single-particle state a legible name. This series encodes an electron's quantum numbers on the four axes of a state quaternion:
Q = (E/Λ)·e₀ + q·e₁ + m_s·e₂ + (ℓ + i m_ℓ)·e₃,
with the energy on the real axis e₀ (Λ = 310 MeV the confinement scale), the electric charge on e₁, the spin projection on e₂, and — a genuinely useful device — the orbital angular momentum on the complex axis e₃, carrying ℓ in its real part and m_ℓ in its imaginary part, so that the three 2p states (m_ℓ = −1, 0, +1) are held apart as 1−i, 1, 1+i. This is a compact, faithful identity card for the state we will call |α⟩,
3The Double Cover Supplies the Exchange Sign
Here is the quaternion's real contribution. The rotations of ordinary space form the group SO(3); the unit quaternions form SU(2), which covers SO(3) twice — every spatial rotation corresponds to two quaternions, q and −q, and a continuous 2π rotation returns the quaternion not to +1 but to −1, with only a 4π turn restoring the identity. A spin-½ particle is an object that lives on this double cover: it is a half-winding of the quaternion sphere. This is not metaphor — it was measured, in 1975, by passing neutrons through a magnetic field and observing that a 2π rotation flips the sign of the beam's interference, while 4π restores it. Now the exchange. By the Finkelstein–Rubinstein theorem, for solitonic objects the operation of exchanging two identical particles is topologically equivalent to rotating one of them by 2π while the other is held fixed. For a double-cover object, that 2π rotation multiplies the state by −1. Therefore: exchange of two identical spin-½ windings ⟹ ×(−1). That is the antisymmetry sign, and it is delivered by the quaternion double cover — the one piece of Pauli that this framework can honestly call its own. (What
4The Tensor Product Turns the Sign into the Rule
With the sign in hand, exclusion is the standard antisymmetry of the multi-particle Hilbert space. Single-particle states are orthonormal vectors |α⟩ in a Hilbert space ℋ; a two-fermion state is the antisymmetrised combination — the Slater determinant, an element of the exterior square Λ²ℋ: |α⟩ ∧ |β⟩ = (|α⟩ ⊗ |β⟩ − |β⟩ ⊗ |α⟩)/√2. The minus sign is exactly the exchange sign of §3, now doing its work. The squared norm is a clean, universal formula,
‖ |α⟩ ∧ |β⟩ ‖² = 1 − |⟨α|β⟩|²,
which vanishes iff |α⟩ and |β⟩ are the same ray and is nonzero for every distinct pair. This is correct for all states — crucially including two states that differ only in energy, such as 1s↑ and 2s↑, which are orthogonal (⟨1s|2s⟩ = 0) and therefore give ‖α∧β‖² = 1, freely allowed, exactly as lithium requires: The operation is a genuine wedge — the antisymmetric ∧ the principle demands — but it is the exterior product of the Hilbert space, acting on state vectors through their inner product, and the quaternion of §2 is the label of the vectors, not the operator. It is worth saying once why a label-product cannot stand in for it: any bilinear product of two four-component labels — the quaternion commutator ½(QP − PQ) included — is a function of the labels alone; but ‖α∧β‖² depends on ⟨α|β⟩, the overlap of the full states, which no product of labels reconstructs. (The commutator specifically equals Im(Q)×Im(P), the cross product of the imaginary parts: it never sees the real/energy axis and vanishes whenever the imaginary parts are parallel, so it would wrongly forbid 1s↑ and 2s↑. pauli_check.py prints that zero, and confirms the tensor-product rule above gets every pair right.)
pauli_check.py — pauli_check
(A) THE BUG: the commutator-wedge is blind to the real (energy) axis
1s_up = (0.001648343206451613, -1, 0.5, 0.0)
2s_up = (0.0016483761241935485, -1, 0.5, 0.0)
wedge(1s_up, 2s_up) = (0.0, 0.0, 0.0, 0.0) |w| = 0.00e+00
-> ZERO: two DISTINCT allowed states are falsely forbidden.
wedge(1s_up, 1s_dn) = (0.0, 0.0, 0.0, 1.0) |w| = 1.000 (He, ok)
Cause: 1/2(QP-PQ) = Im(Q) x Im(P). The real parts never enter;
it vanishes when the imaginary parts are parallel, not only identical.
(B) THE FIX: Slater determinant in the Hilbert space, ||a^b||^2 = 1 - |<a|b>|^2
1s_up vs 2s_up (distinct n, orthogonal) ||a^b||^2 = 1 ALLOWED <- (A) got this WRONG
1s_up vs 1s_up (same ray) ||a^b||^2 = 0 FORBIDDEN
1s_up vs 1s_dn (orthogonal spin) ||a^b||^2 = 1 ALLOWED
1s_up vs 2p_up (orthogonal l) ||a^b||^2 = 1 ALLOWED
Correct for EVERY pair. The exclusion is a tensor-product antisymmetry;
the quaternion may LABEL |a>, but it does not PERFORM the antisymmetrisation.
# -*- coding: utf-8 -*-
"""Pauli exclusion, diagnosed and repaired. Pure standard library.
Demonstrates:
(A) THE BUG in the State Quaternion / Pauli papers: exclusion defined as the
quaternion 'wedge' Q^P = 1/2(QP - PQ) is the CROSS PRODUCT of the imaginary
parts, Im(Q) x Im(P). It is blind to the real (energy) axis e0 and vanishes
whenever the imaginary parts are parallel. So 1s-up and 2s-up -- distinct
allowed states that differ ONLY in energy -- give Q^Q = 0 and are FALSELY
excluded, exactly the lithium problem the State paper worried about.
(B) THE FIX: antisymmetrise in the Hilbert space. Single-particle states are
orthonormal vectors |a> labelled by full quantum numbers; the two-fermion
state is the Slater determinant |a>^|b> in Lambda^2(H), with
|| |a>^|b> ||^2 = 1 - |<a|b>|^2 : zero IFF same ray, nonzero for every
distinct pair. Correct for all pairs, including the ones (A) gets wrong.
"""
# --- quaternion algebra (stdlib) ------------------------------------------
def qmul(a, b):
a0,a1,a2,a3 = a; b0,b1,b2,b3 = b
return (a0*b0 - a1*b1 - a2*b2 - a3*b3,
a0*b1 + a1*b0 + a2*b3 - a3*b2,
a0*b2 - a1*b3 + a2*b0 + a3*b1,
a0*b3 + a1*b2 - a2*b1 + a3*b0)
def wedge(Q, P): # the State paper's definition
QP = qmul(Q, P); PQ = qmul(P, Q)
return tuple(0.5*(QP[i]-PQ[i]) for i in range(4))
def norm(v): return sum(x*x for x in v)**0.5
# --- state quaternions [E/Lambda, q, m_s, L] -----------------------------
Lam = 310.0
Q1s_up = ((0.511-13.606e-6)/Lam, -1, +0.5, 0.0)
Q2s_up = ((0.511-13.606e-6/4)/Lam, -1, +0.5, 0.0) # differs from 1s ONLY in energy
Q1s_dn = ((0.511-13.606e-6)/Lam, -1, -0.5, 0.0) # differs in spin
print("(A) THE BUG: the commutator-wedge is blind to the real (energy) axis\n")
print(f" 1s_up = {Q1s_up}")
print(f" 2s_up = {Q2s_up}")
w = wedge(Q1s_up, Q2s_up)
print(f" wedge(1s_up, 2s_up) = {tuple(round(x,6) for x in w)} |w| = {norm(w):.2e}")
print(" -> ZERO: two DISTINCT allowed states are falsely forbidden.")
w2 = wedge(Q1s_up, Q1s_dn)
print(f" wedge(1s_up, 1s_dn) = {tuple(round(x,6) for x in w2)} |w| = {norm(w2):.3f} (He, ok)")
print(" Cause: 1/2(QP-PQ) = Im(Q) x Im(P). The real parts never enter;")
print(" it vanishes when the imaginary parts are parallel, not only identical.\n")
print("(B) THE FIX: Slater determinant in the Hilbert space, ||a^b||^2 = 1 - |<a|b>|^2\n")
def slater(overlap): return 1 - abs(overlap)**2
cases = [("1s_up vs 2s_up (distinct n, orthogonal)", 0.0, "ALLOWED <- (A) got this WRONG"),
("1s_up vs 1s_up (same ray)", 1.0, "FORBIDDEN"),
("1s_up vs 1s_dn (orthogonal spin)", 0.0, "ALLOWED"),
("1s_up vs 2p_up (orthogonal l)", 0.0, "ALLOWED")]
for name, ov, verdict in cases:
print(f" {name:42s} ||a^b||^2 = {slater(ov):.0f} {verdict}")
print("\n Correct for EVERY pair. The exclusion is a tensor-product antisymmetry;")
print(" the quaternion may LABEL |a>, but it does not PERFORM the antisymmetrisation.")
5Filling the Table
The two pieces together generate atomic structure with no further postulate. Each shell is a finite set of orthonormal labels |α⟩; the exclusion rule permits any antisymmetrised combination and forbids repetition of a ray; so a shell of k distinct states holds exactly k electrons, and the (k+1)-th must open the next shell. The n = 2 shell has eight labels (2s and three 2p, each with two spins), all mutually orthogonal, hence all mutually allowed and jointly filled; a ninth electron would have to duplicate a ray, giving ‖α∧α‖² = 0. The periodic table's period lengths 2, 8, 8, 18, … are the counts of orthonormal labels per shell — the quaternion labels them, the tensor product excludes duplicates, and chemistry follows.
6Why a Single Quaternion Cannot Carry the Rule — Stated Once
For the record, and so the design choice is not repeated: exclusion cannot be a binary operation on two state labels producing a number. The reason is structural, not a matter of choosing a better product. The rule must vanish for identical states and only for identical states, i.e. it must be a function of the overlap ⟨α|β⟩ of the full Hilbert-space vectors. A product of four-component labels is a function of the labels, and the label is a lossy summary — two orthogonal states (1s, 2s) can carry parallel imaginary parts, and any product blind to the real axis (the commutator is exactly such) collapses them. The antisymmetrisation therefore must be performed in the Hilbert space, where the inner product lives; the quaternion's job ends at naming the vectors.
7What Is Quaternionic, What Is Standard, What Is Owed
Quaternionic (claimed). The exchange sign −1, from
8The Sentence
Exclusion is a sign and a rule with two owners: the quaternion double cover, half-wound and measured in 1975, hands over the minus sign of exchange; the Hilbert-space tensor product turns that minus into the law that no two electrons may share a ray; and the quaternion, having given the sign, spends the rest of its service merely as the name on each electron's identity card.
References
W. Pauli, Z. Phys. 31, 765 (1925); M. Fierz, Helv. Phys. Acta 12, 3 (1939) and W. Pauli, Phys. Rev. 58, 716 (1940) — spin–statistics; D. Finkelstein and J. Rubinstein, J. Math. Phys. 9, 1762 (1968) — exchange as rotation; H. Rauch et al., Phys. Lett. A 54, 425 (1975) — the 4π neutron interferometry; standard treatments of the Slater determinant and the exterior algebra of Hilbert space; and the notes of this series (the State Quaternion label; the Three Storeys — the seal and the facade). Verification script: pauli_check.py (stdlib). This edition supersedes the earlier draft that defined exclusion via the quaternion commutator. (Citations from memory; the literature-verification pass applies.) Acknowledgment: drafting and numerical checks by machine (Claude, Anthropic), in conversation; the instinct that exclusion should wear an antisymmetric wedge and that fermion statistics is