Belousov–Zhabotinsky

To Write Boris Belousov: pharmacologist, not chemist. Studying the Krebs cycle by analogy. Mixed bromate, cerium, malonic acid, and citric acid in dilute sulfuric acid solution. The solution turned yellow (Ce⁴⁺, cerium(IV)), then cleared (Ce³⁺), then yellow again. Not once: repeatedly, for minutes, then hours, at regular intervals of about 25 seconds. He knew immediately this was a non-equilibrium oscillation sustained by coupled autocatalytic reactions. Reviewers said oscillating reactions were thermodynamically impossible. They confused a closed-system equilibrium argument with an open-system trajectory argument: BZ is not at equilibrium, it is on a limit cycle far from equilibrium, dissipating free energy. The system is not a perpetual motion machine; it is a clock that winds down gradually. Belousov's insight was correct. He never got credit in his lifetime. Died in 1970. Prigogine–Defay award (posthumous recognition) came later; Prigogine's Nobel (1977) for dissipative structures depends on the same physics. The BZ reaction is one of the intellectual crimes of 20th-century science.

To Write Anatol Zhabotinsky (Moscow State University, 1961): reproduced and extended Belousov's work. Key addition: Ferroin (iron-phenanthroline complex) as indicator, giving vivid red/blue colour change visible across a room. Thin-layer BZ: poured into petri dish, the solution forms concentric target waves (circular expanding rings) and spiral waves (Archimedean spirals). Art Winfree (1972, Science): the topological theory. Each spiral has a phase singularity at its tip: a point where the phase of the oscillation is undefined. This is a topological defect. The spiral cannot be removed by smooth deformation. In dm³: a phase singularity is a Legendrian singularity on the contact manifold (M, α). The spiral wave is a Legendrian knot. The tip is where the contact form α = 0. Count the spirals: they come in pairs of opposite chirality (left-hand and right-hand), consistent with the topological charge ±1. Winfree called this “the geometry of biological time”; dm³ adds: the geometry is contact geometry.

To Write The Oregonator reduces to a 3-variable system near the limit cycle. In the slow manifold approximation it becomes 2D: the FitzHugh-Nagumo form, which is also the neuron model. The BZ reaction and the action potential are the same mathematical object at different scales. dm³ observation: the π operator sets the period T* = 2π; the Oregonator limit cycle has period T ≈ 25 seconds in vitro. The correspondence is not numerical (units differ) but structural: same topology, same contact-geometric Reeb flow, same Lyapunov exponent μ = −2 for perturbation decay.

To Write Jean-Claude Roux (Bordeaux, 1983): experimental confirmation of Feigenbaum period-doubling cascade in the BZ reaction. As a control parameter (flow rate in CSTR) is increased, the BZ oscillation goes through a sequence of period doublings: T → 2T → 4T → 8T → chaos. The ratio of successive bifurcation intervals converges to δ = 4.6692... (Feigenbaum constant). This was the first experimental measurement of Feigenbaum universality in a real chemical system. The significance: the Feigenbaum constant δ is universal — it appears in every unimodal period-doubling cascade regardless of the specific equations. Just as π and φ are universal constants that appear independently of the particular problem, δ is a universal constant of the transition to chaos. The dm³ recurrence ladder reads this sequence in reverse: Omega → Sigma → Delta → eta → phi → pi is the anti-period-doubling direction, converging to order rather than chaos. Cross-reference: ch-lorenz-chaos.html (§4 on Feigenbaum and the n-bonacci antidote).

To Write The BZ reaction is the laboratory proof of concept. The same physics — autocatalytic feedback, inhibition, spatial diffusion, emergent periodicity — appears at every scale of physical reality:

To Write The claim: the operator chain G = U ∘ F ∘ K ∘ C is not a description of one system. It is the universal structure that any dissipative system with feedback and diffusion satisfies. BZ is the laboratory prototype. Neural circuits are the biological implementation. Galaxy formation is the cosmological instance. The constants π, φ, μ, η, Δ, Σ, Ω appear in each domain not because we put them there but because they are the eigenvalues of the contact-geometric constraint. This is the claim of the dm³ programme. It requires proof at each scale. The book is the proof.

To Write The BZ reaction was misunderstood for 20 years partly because no one could simulate it. Once the Oregonator was in a computer (1972), the limit cycle was immediately visible: the trajectory spirals toward the attractor and stays there. The computer did not prove the chemistry; it made the geometry legible. This is the role of computation in complex systems science: not to replace mathematical proof but to make the attractor visible before the proof exists. AXLE plays this role for dm³: it mechanises the geometry, makes the operator chain executable, lets the theorem-finder see what the limit cycle looks like before the Lean proof closes. The two Lean theorems already proved (critDim_monotone, no_return_to_critical) were found this way: simulation first, proof second. The remaining sorry-stubs (turing_instability, G_stabilises_turing_pattern, nBonacci_is_Hausdorff_dim, the Feigenbaum duality) are the next targets. The computer shows us where to look. The mathematics confirms what we see.