Virtual Monopoles in the FeMo-Cofactor: An Accessible Guide

Plain-language explainer for doi:10.5281/zenodo.20346650 (#318)

The central idea in one sentence

The iron-molybdenum cofactor of nitrogenase — the enzyme that fixes atmospheric nitrogen into the molecules of life — has resisted explanation for fifty years, and we propose that three of its deepest anomalies share a single topological cause: its seven-atom iron-carbide core realises the Fano plane, the simplest finite projective geometry, whose symmetry group is the exceptional Lie group $G_2$.

The mystery: the enzyme that shouldn’t work as well as it does

Every living organism that uses nitrogen — which is every living organism — ultimately depends on nitrogenase, the enzyme that breaks the triple bond of atmospheric N₂ and converts it into ammonia. Without nitrogenase, there is no fixed nitrogen; without fixed nitrogen, there are no amino acids, no DNA, no life.

Nitrogenase does this at room temperature and atmospheric pressure, using a metal cluster called the FeMo-cofactor. Industrial nitrogen fixation (the Haber-Bosch process) requires 400°C and 200 atmospheres of pressure. The enzyme is, by any measure, a chemical miracle.

Three anomalies have resisted fifty years of explanation:

Anomaly 1: DFT failure. Broken-symmetry density functional theory — the standard computational tool for open-shell metal clusters — consistently mis-assigns the spin states of the individual iron atoms in the resting state. The ground state has spin $S = 3/2$ by EPR spectroscopy, but DFT cannot reproduce this without imposing an artificial, symmetry-broken initial guess. The error is systematic, not numerical: the functional form is wrong.

Anomaly 2: The silent carbide. In 2011 it was discovered that the centre of the FeMo-cofactor contains a single carbon atom, a $\mu_6$-carbide, coordinated to all six iron atoms simultaneously. It is essential for activity — delete it and the enzyme is dead — yet it undergoes no detectable chemistry during the catalytic cycle. It is not protonated, not reduced, not exchanged. Chemists have no explanation for why something chemically inert is functionally indispensable.

Anomaly 3: The ATP overhead. Nitrogenase consumes 16 ATP per N₂ fixed. The thermodynamic minimum for breaking the N≡N triple bond requires only 4 ATP. The four-fold excess has no accepted explanation. The energy is not wasted as heat in any obvious way — it simply disappears into the mechanism.

The proposal: a Fano plane in your enzyme

The FeMo-cofactor core consists of six iron atoms arranged in a trigonal prism, with the central carbide coordinating all six. Count the atoms: six irons plus one carbide equals seven.

The Fano plane is the simplest finite projective geometry: seven points and seven lines, with exactly three points on every line and exactly three lines through every point. Its automorphism group — the group of symmetries that map it to itself — is the exceptional Lie group $G_2$, the same group that describes the algebra of the octonions.

We propose that the seven-atom Fe₆C core realises the Fano plane: the exchange interactions between atoms follow the Fano incidence structure. If true, the electronic Hamiltonian has $G_2$ symmetry, not the lower crystallographic $D_3$ symmetry that the geometric arrangement of atoms suggests.

This single hypothesis addresses all three anomalies simultaneously:

Anomaly 1 resolved. A $G_2$-symmetric Heisenberg model with two exchange constants — antiferromagnetic between iron atoms, weaker antiferromagnetic through the carbide — reproduces $S = 3/2$ exactly, without broken symmetry, at the physical spin value $S = 2$ per iron site (Hilbert space dimension $5^7 = 78{,}125$, solved by sparse Lanczos diagonalisation). DFT fails because it imposes $D_3$ symmetry and then breaks it; the correct symmetry is $G_2$ and should not be broken.

Anomaly 2 resolved. The carbide is not chemically active but topologically essential: it is the seventh point of the Fano plane. Remove it and the Fano structure — and with it the $G_2$ symmetry and the $S = 3/2$ ground state — collapses. The carbide’s role is geometric, not chemical.

Anomaly 3 resolved. The excess ATP cost corresponds, within 15%, to the energy cost of the Fano-line closure operation (see below). The enzyme pays a topological tax.

A second Fano plane: the nitrogen molecule itself

Remarkably, the frontier molecular orbitals of N₂ — the seven quantum states most directly involved in bonding — also realise a copy of the Fano plane:

\[\sigma,\; \pi_x,\; \pi_y,\; \pi^{*}_x,\; \pi^{*}_y,\; \sigma^*,\; \mathrm{N_2}\]

Five of the seven Fano lines in this orbital Fano plane couple bonding orbitals to antibonding orbitals. These five lines encode the Dewar–Chatt–Duncanson (DCD) mechanism — the standard chemical explanation for how transition metals activate N₂, via $\sigma$-donation and $\pi^{}$-backdonation — as a *geometric identity of the Fano plane rather than a perturbative approximation. The DCD mechanism is not an approximation that happens to work: it is exact, because it is a consequence of the incidence geometry.

Bond breaking as non-associative geometry

When N₂ binds to the FeMo-cofactor at the E4H4 state, the two Fano planes — the Fe₆C Fano and the N₂ orbital Fano — are glued together by a lattice surgery operation, creating a twelve-node junction. The iron atoms Fe2 and Fe6 play the roles of $\sigma$-donation and $\pi^{*}$-backdonation respectively.

Bond cleavage — the breaking of the N≡N triple bond — is identified with the Fano-Line Closure operation: a mathematical transformation that acts on the Fano line $(\sigma, \pi_x, \pi^{}_x)$ and simultaneously annihilates both the bonding $\sigma$ amplitude and the antibonding $\pi^{}_x$ amplitude in a single step. This operation is non-associative: it cannot be factored into a sequence of ordinary (associative) steps. This is why it is so hard to reproduce computationally and why it requires unusual catalytic machinery. Non-associativity is not a complication to be managed — it is the mechanism.

The full catalytic cycle as a computer program

One of the more striking results of this paper is that the complete Lowe–Thorneley kinetic cycle for biological nitrogen fixation — eight electron transfers, eight proton transfers, N₂ binding, bond cleavage, and product release — can be written as a fifteen-instruction program in the 731-RPU Origami instruction set architecture:

Step	Instruction	Chemical meaning
×4	`FLIP`	Add e⁻/H⁺ to Fe cluster
×1	`SPLIT`	N₂ binds; create 12-node junction
×4	`FLIP`	Add e⁻/H⁺ to junction
×1	`SPLAT`	N≡N → N+N via Fano-Line Closure
×2	`FLIP`	Protonate each N fragment → NH₃
×3	`FLOP`	Release NH₃, NH₃, H₂

Each instruction corresponds to a Pachner move — a local retriangulation of a simplicial complex. The enzyme is, in this sense, a biological implementation of the 731-RPU hardware architecture. The 16 ATP molecules are the power supply.

What would prove this?

The conjectures are falsifiable. Three tests require no new experiments:

Mössbauer reanalysis. The four inequivalent iron environments observed by Mössbauer spectroscopy should correspond to two Fano classes: Fe2/Fe4 (positive moment, $\langle S_z\rangle \approx +0.26$) versus Fe3/Fe5/Fe6/Fe7 (negative moment, $\langle S_z\rangle \approx -0.09$ to $-0.12$). Reanalysis of published hyperfine field data in the $G_2$ site basis would confirm or refute this immediately.
EPR fine structure. The $G_2$ symmetry predicts specific selection rules for the EPR spectrum that differ from those implied by $D_3$. These can be checked against existing data.
DFT with $G_2$ constraint. A DFT calculation that enforces $G_2$ symmetry as a constraint (rather than $D_3$) should converge to a qualitatively different electronic structure, with $S = 3/2$ as the unconstrained ground state.

Two tests require new experiments:

Muon spin relaxation ($\mu$SR). Spin fluctuations consistent with geometric frustration would confirm the molecular frustrated-magnet picture.
Inelastic neutron scattering. Low-energy magnetic excitations below 10 meV, consistent with frustrated exchange, would directly support the conjecture.

Why the stakes are high

If the $G_2$ conjecture is correct, the implications extend well beyond nitrogenase:

It would explain the 50-year failure of DFT for this system: the wrong symmetry was assumed from the start.
It would identify the carbide’s role unambiguously, after a decade of speculation since its discovery.
It would connect biological catalysis to the mathematics of exceptional Lie groups and octonions — a connection that has no precedent.
It would suggest a design principle for synthetic catalysts: build the Fano geometry in, and the $G_2$ symmetry will do the rest.
It would imply that non-associative algebra — long considered a mathematical curiosity — plays a functional role in the chemistry of life.

These are large claims, stated here as conjectures. The calculations support them; the experimental tests will decide.