The Periodic Table Is a Theorem — Not a Rule

Plain-language explainer for doi:10.5281/zenodo.21479618 (#652)

The problem with the Madelung rule

Every chemistry student memorises the Madelung rule: fill orbitals in order of increasing n+ℓ, where n is the principal quantum number and ℓ is the angular momentum. This gives the sequence 1s → 2s → 2p → 3s → 3p → 4s → 3d → 4p → … and correctly predicts the structure of the periodic table for the first 118 elements (with a handful of well-known exceptions).

The problem: no one knows why it works. The Madelung rule is not derived from the Schrödinger equation. It is an empirical observation dressed up as a rule. Every textbook that presents it as fundamental is quietly sweeping an open question under the rug.

This paper derives it from first principles — from the geometry of twistor space.

What twistor space is

Twistor space CP³ is a mathematical space introduced by Roger Penrose in 1967 as the natural arena for combining quantum mechanics with spacetime geometry. It is a complex projective 3-space: four complex dimensions, with the overall complex scale identified (so effectively six real dimensions), and with a rich internal structure inherited from the conformal symmetry of 4-dimensional spacetime.

The key fact for chemistry: the symmetry group of CP³ is SU(2,2), which is the same group as SO(4,2) — the conformal group of 4D Minkowski spacetime. And SO(4,2) is precisely the dynamical symmetry group of the hydrogen atom, first identified by Barut and Bohm in 1965 using the spectrum-generating algebra approach.

In other words: the symmetry that organises twistor space is the same symmetry that organises hydrogen’s energy levels. This is not a coincidence — it is the heart of the paper.

Hydrogen orbitals as line bundles on CP³

In algebraic geometry, a line bundle O(k) on CP³ is a family of complex lines (one-dimensional complex vector spaces) attached to each point of CP³ in a way that twists k times as you go around. Different values of k give qualitatively different bundles.

The Penrose transform — a deep result connecting twistor geometry to physics — maps:

  • Holomorphic sections of O(n−1) on CP³ ↔ hydrogen wavefunctions at energy level n

More precisely: the wavefunction ψ_{n,ℓ,m} for orbital (n, ℓ, m) is an element of H⁰(CP³, O(n−1)) — a holomorphic section of the line bundle of degree n−1. The angular quantum numbers (ℓ, m) label which SO(3)-subrepresentation it belongs to within the n²-dimensional space of degree-(n−1) sections.

The principal quantum number n is the line bundle degree plus one. The entire shell structure of hydrogen — the fact that shell n contains n² states — is the statement that O(n−1) has exactly n² independent holomorphic sections on CP³.

The one-line proof of Madelung

The Picard group of a space is the group of all line bundles on it, under the operation of tensor product. For CP³:

\[\text{Pic}(\mathbb{CP}^3) = \mathbb{Z}\]

generated by O(1). Every line bundle is O(k) for some integer k, and there is exactly one natural total ordering on the integers: the standard ordering k = 0, 1, 2, 3, …

Now: the Madelung diagonal n+ℓ indexes which line bundle degree controls the orbital. Specifically, n+ℓ = k+1 where O(k) is the relevant bundle degree. Filling orbitals in order of increasing n+ℓ is the same as filling line bundles in order of increasing degree.

There is only one natural ordering on Pic(CP³) = ℤ. Therefore the Madelung rule is the unique natural filling order forced by the twistor geometry. It is not a choice or an empirical observation — it is the only order consistent with the projective structure of CP³.

What this explains about Madelung anomalies

The well-known exceptions to the Madelung rule — Cr (prefers 3d⁵4s¹ over 3d⁴4s²), Cu (prefers 3d¹⁰4s¹ over 3d⁹4s²), and about a dozen others among transition metals and lanthanides — are often dismissed as “exchange energy effects.” This paper gives a geometric interpretation.

Near the boundary between Madelung diagonals (where two different line bundles O(k) and O(k+1) are nearly degenerate in energy), small perturbations — relativistic corrections, electron-electron repulsion, spin-orbit coupling — can tip the balance. The anomalies are not failures of the rule; they are near-degenerate line bundle crossings at the projection window boundary.

Quantitative prediction: the energy splitting between the anomalous configuration and the Madelung prediction should be proportional to the distance of the orbital from the relevant Madelung diagonal in (n, ℓ) space. This is testable from spectroscopic data.

The ISA connection: sheaf cohomology as H^k

The Origami ISA identifies three tiers of quantum computation:

  • H⁰ (ORBIT): classical, tropical, stabiliser-level
  • H¹ (TWIST): Clifford, topological, surface-code level
  • H² (BIND): metaplectic, genuinely quantum, colour-code level

In the twistor picture, these tiers correspond to sheaf cohomology groups of CP³:

  • H⁰(CP³, O(k)): holomorphic sections = bound state wavefunctions (chemistry)
  • H¹(CP³, O(k)): first cohomology = massless free fields (quantum field theory)
  • H²(CP³, O(k)): second cohomology = sources and interactions (interacting QFT)

The H^k ladder of the ISA is therefore not just an analogy with cohomology — it is sheaf cohomology on CP³, realised physically at three different scales: chemistry (H⁰), free QFT (H¹), and interacting QFT (H²).

This gives the ISA a geometric foundation in twistor theory and provides a unified language spanning from atomic orbitals to quantum field theory.

The β-duality between two approaches

Paper 652 also identifies a striking duality:

  • The amplituhedron approach to scattering amplitudes works from the outside in: it starts with the full Grassmannian Gr(k, n) and takes residues to extract physical amplitudes.
  • The CASSCF approach to quantum chemistry works from the inside out: it starts with a small active space and expands outward to capture correlation.

These two procedures are related by β-duality — the MGE parameter β that interpolates between the tropical (classical, β→∞) limit and the quantum (β→0) limit. At β→∞, CASSCF contracts to the active space minimum; the amplituhedron contracts to the leading singularity. They meet at the same geometric object: the positive Grassmannian.

Chemistry and particle physics are computing the same twistor integrals from opposite ends of the β-line.

What this paper does not claim

  • It does not derive the Madelung anomalies quantitatively — it explains their geometry but does not compute their magnitude from first principles.
  • It does not replace the Schrödinger equation — the twistor picture is a reformulation, not a replacement. Every result here is consistent with standard quantum mechanics.
  • The Penrose transform identification of ψ_{n,ℓ,m} with sections of O(n−1) is rigorous for the free hydrogen atom; extension to many-electron atoms requires further work (Paper 653 begins this).

This is Part I of a trilogy

  • Paper 652 (this paper): Madelung rule from Pic(CP³) = ℤ; hydrogen orbitals as line bundle sections; β-duality amplituhedron↔CASSCF
  • Paper 653: ISA opcodes as sheaf cohomology; H^k = H^k(CP³, O(k)); the full ISA-twistor dictionary (doi:10.5281/zenodo.21479624)
  • Paper 654: Collaborative work with L.P. Hughston — the Madelung fluid as the real part of a holomorphic flow on CP³; quantum hydrodynamics from twistor geometry (in preparation)

See also:

For the full technical treatment, see doi:10.5281/zenodo.21479618