NOTE-414: Belousov-Zhabotinsky reaction

Contribution

An encyclopedia review, so no new result. What it adds to the record is a first-hand account, by one of the reaction’s two eponyms, of:

• the history: Belousov’s citric-acid/cerium system and Zhabotinsky’s malonic-acid version;
• the core mechanism, the Oregonator and the Oregonator’s known failures;
• the difference between closed and open-reactor dynamics;
• the wave phenomena first reported by Zaikin and Zhabotinsky in 1970.

Key insight

BZ is an oscillator because a fast positive feedback (autocatalytic HBrO₂ production oxidizing the catalyst) is coupled to a slower negative feedback (Br⁻ released as the oxidized catalyst is reduced by the organic substrate). Each switches the other off at a threshold, which gives a relaxation limit cycle. The same thresholds make the unstirred medium excitable, so a local switch propagates as a trigger wave followed by a refractory zone. The rest follows from that: annihilation of colliding waves, the fastest pacemaker entraining the others, and spirals at broken fronts.

Assumptions

What the review takes for granted:

• Thermodynamics. Oscillation needs the system “sufficiently far from the state of thermodynamic equilibrium”, and detailed balance forbids it near equilibrium (both cited to Nicolis and Prigogine 1977, LIT-524).
• Chemistry.
  • Bromate is “the only irreplaceable initial reagent”.
  • The catalyst may be Ce, Mn or complexes of Fe, Ru, Co, Cu, Cr, Ag, Ni and Os.
  • Many reductants work.
• Models. The Oregonator (Field and Noyes 1974) and Tyson’s two-variable reductions are taken as qualitative models. The quantitative claims rest on later modified versions (Rovinsky and Zhabotinsky 1984; Nagy-Ungvarai et al. 1989).

Key results

• History.
  • Belousov found oscillations with Ce³⁺/Ce⁴⁺ and citric acid, with frequency rising with temperature (Belousov 1959).
  • Zhabotinsky replaced citric acid with malonic acid and showed that the colour oscillation is the oscillation of [Ce⁴⁺]. He found that oxidation of Ce³⁺ by HBrO₃ is autocatalytic, that oscillation starts after bromomalonic acid accumulates, and that Br⁻ inhibits the autocatalysis (Zhabotinsky 1964a, b).
• Core scheme (Vavilin and Zaikin 1971):
  • HBrO₃ + HBrO₂ → 2BrO₂• + H₂O;
  • H⁺ + BrO₂• + Fe(phen)₃²⁺ → Fe(phen)₃³⁺ + HBrO₂;
  • HBrO₂ + H⁺ + Br⁻ → 2HOBr;
  • plus Br⁻ production as the oxidized catalyst oxidizes bromomalonic acid. Phase resetting (Fig. 1, Vavilin et al. 1973) supports it. Br⁻ or Ce⁴⁺ injected while [Ce⁴⁺] is rising switches the system to the falling phase. Ag⁺, which removes Br⁻, switches it from falling to rising.
• Oregonator. Field, Kőrös and Noyes (1972) gave the detailed mechanism. The Oregonator (Field and Noyes 1974) has variables [HBrO₂], [Br⁻] and [Ce⁴⁺] and a stoichiometric factor standing for bromide production. It “properly models oscillations and excitability” but “is not a quantitative model”: it leaves total catalyst out of its parameters, reproduces the waveform poorly, and misses the observed oscillatory domains. Taking reversibility of the catalyst reactions into account corrects these.
• Closed systems.
  • Oscillation occurs over large ellipsoidal domains of initial concentrations, spanning about three orders of magnitude in bromate and malonic acid and three to four in cerium.
  • Period-2 oscillation can arise as the reaction evolves.
  • Excitability and bistability lie outside the oscillatory domain. Bistability needs a source of Br⁻ other than the reactions of Ce⁴⁺.
  • A closed system “can generate up to several thousand oscillatory cycles”.
• Open systems. In a CSTR “stationary oscillations can continue indefinitely”. Bursting (Vavilin et al. 1968) and chaos (Schmitz et al. 1977) were found, and the CSTR allows bifurcation sequences to be traced.
• Waves.
  • Zaikin and Zhabotinsky (1970) observed concentric trigger waves from point pacemakers in a thin unstirred ferroin-catalysed layer, forming target patterns.
  • Colliding waves annihilate because of their refractory zones, and the fastest pacemaker entrains the rest (Fig. 4).
  • Broken fronts curl into spirals (Zhabotinsky and Zaikin 1971; Winfree 1972), and in thick layers into scroll waves (Winfree 1973).
  • With 1,4-cyclohexanedione, anomalous dispersion groups waves into packets.
  • In AOT microemulsion BZ shows Turing and short-wave instabilities and “produces Turing structures”, standing waves, localized structures, inwardly moving waves and segmented waves (Vanag and Epstein 2001).

Claims

id	claim	strength	support
C1	Before BZ, chemists generally attributed homogeneous oscillations to heterogeneity or error, misreading the near-equilibrium prohibition as general	historical assertion	Zhabotinsky 1991 (his own history)
C2	The oscillation is driven by autocatalytic HBrO₂ inhibited by Br⁻, with Br⁻ regenerated from bromomalonic acid by the oxidized catalyst	strong	primary kinetics cited; phase-resetting experiments (Fig. 1)
C3	The Oregonator models oscillation and excitability qualitatively but not quantitatively	strong	named defects; corrected models cited
C4	A closed BZ system can run up to several thousand cycles	stated, uncited	§Basics
C5	In a CSTR oscillation continues indefinitely, with bursting and chaos in some regimes	strong	cited experiments
C6	Unstirred BZ supports target, spiral and scroll waves, which annihilate on collision	strong	Zaikin and Zhabotinsky 1970; Winfree 1972, 1973
C7	BZ in microemulsion produces Turing structures	moderate here (one cited group)	Vanag and Epstein 2001

Concepts

• BZ reaction: “a family of oscillating chemical reactions” in which transition-metal ions catalyse oxidation of (usually organic) reductants by bromic acid in acidic water.
• trigger wave: a pulse of excitation followed by a refractory zone, propagating through an excitable or oscillatory medium.
• pacemaker: a point source emitting concentric waves; the fastest one entrains the medium.
• CSTR: continuous-flow stirred tank reactor, which keeps reagent concentrations fixed by inflow and outflow.

Connections

• Prigogine’s Nobel lecture (LIT-516), read. The lecture does not name BZ; its “chemical clock” is the Brusselator, a model with A and B “maintained constant”. BZ in a CSTR is the experimental counterpart of that idealization. A closed BZ batch is not: there A and B run down, and the oscillation is a transient on the way to equilibrium. Both share the lecture’s necessary condition, an autocatalytic step.
• Turing (LIT-535), read. BZ waves are the chemical travelling waves Turing’s case (e) gestured at, but excitable and strongly non-linear, which his linear theory does not cover. The review’s last paragraph records that BZ in microemulsion has since given stationary Turing structures.
• Cross and Hohenberg (LIT-527), skimmed. Their §X uses BZ and the Oregonator as the chemical example and says the same things: closed experiments “run down in finite time (typically less than 100 periods of oscillation)” for spatial patterns, and open gel reactors were the remedy. Their figure of under 100 periods for unstirred pattern experiments and the review’s “several thousand” cycles for closed systems are not in conflict: the setups differ.
• Nicolis and Prigogine (LIT-524), Deferred. Cited by the review for the far-from-equilibrium condition on oscillation.

Bearing on the record

• LIT-192 (demarcation). BZ has rhythm, excitability, self-propagating signals and entrainment, which are properties often taken as marks of living organization. It has no closure of constraints and no self-maintenance: its organization is maintained by the experimenter’s feed. That makes it the cleanest chemical foil for the candle-flame-versus-organism line Nahas and Sachs discuss (NOTE-094). It shows that a criterion based on thermodynamics or dynamics alone does not separate organisms from non-organisms, so whatever separates them must be organizational.
• LIT-211 (the “Dissipative Self-Organizing Systems” cluster). BZ in a CSTR satisfies definitions of that kind as well as an organism does. That is the standing objection to them, and NOTE-109 records that the cluster’s definitions are “persister” criteria of the kind Baedke criticizes (LIT-167). BZ is a persister only while fed.
• No THEORY is indicated by this reading alone. No ML instruction.

Note of 2026-10-16: THEORY-508 now cites LIT-531 as a source.

Limitations

• A short review. No derivations, and some statements (such as “several thousand” cycles) are uncited.
• Authorship is mixed. The page credits Zhabotinsky but lists later contributors and was modified in 2011.
• Nothing on thermodynamics beyond the opening paragraph. No entropy production, free-energy budget or stability analysis is given.

Open questions

• How long does a closed BZ batch stay effectively at a quasi-steady state? That determines whether its patterns count as dissipative structures in Prigogine’s steady-state sense or only as transients.
• Do the quantitative corrections to the Oregonator change any of its qualitative wave predictions?

Corrections

• none to a seeded skim (there was no seed)
• Request’s candidates. The request’s alternative citation, Zaikin & Zhabotinsky (1970), Nature 225:535–537, is right; the issue is 5232 and the DOI 10.1038/225535b0. The first author is A. N. Zaikin. Nature prints the title in title case: “Concentration Wave Propagation in Two-dimensional Liquid-phase Self-oscillating System”.
• Discovery date. Cross and Hohenberg (§X.A.1) date Belousov’s discovery to 1951; the review cites its first publication, Belousov 1959. Both can be right; neither source reconciles them.