Flow Lab

Syllabus / Module 15: Applications: Biology, Chemistry & Society

Lesson 15.4

Chemical Kinetics & Stiff Systems

Mass-action kinetics, reaction networks, Michaelis–Menten, the Robertson problem, and oscillating reactions (Brusselator, BZ).

Written byClaude Sonnet

🌊
Why this matters for simulation: Combustion simulation couples stiff chemistry to fluid flow.

By the end of this lesson you can…

  • Turn a reaction network into ODEs by mass action and read off its conserved sums.
  • Use integrated rate laws and half-lives, and solve a reversible reaction to its equilibrium.
  • Solve the chain A to B to C exactly and say when the quasi-steady approximation for B is accurate.
  • Derive the Michaelis-Menten rate and test its validity with $\varepsilon = E_0/(K_m + S_0)$.
  • Find the Brusselator's Hopf threshold $b = 1 + a^2$ and its frequency at onset.
  • Explain stiffness with a measured step-size comparison of RK4 against implicit methods.
On this page
·
  1. Mass action: networks as ODEs
  2. Reversible reactions and equilibrium
  3. Consecutive reactions and the quasi-steady state
  4. Michaelis-Menten kinetics
  5. Oscillating reactions: the Brusselator
  6. Stiffness, measured
  7. Where this goes next
  8. The exercise: Michaelis-Menten, Lindemann, backward Euler and the Brusselator
  9. Concept checks
  10. Problems
  11. Key Takeaways

15.3 Epidemic Models ended on a coincidence that is no coincidence. The SIR product S IS\,I is the law of mass action of chemistry: molecules react at a rate proportional to how often they meet, and meetings are proportional to the product of concentrations. Rate constants here differ by factors of a thousand or a billion, one reaction races ahead and finishes, and an explicit solver pays for the race for the rest of the run. That is stiffness.

▶ Video Mass action and rate laws, reversible reactions, the A to B to C chain and its quasi-steady state, Michaelis-Menten, the Brusselator oscillator, and a measured step-size comparison between RK4 and implicit methodscaptions only
Transcript
  1. Last lesson ended on a coincidence that is no coincidence. The product S times I in SIR is the law of mass action: molecules react at a rate proportional to how often they meet, so A plus B making C runs at k times A times B. A network becomes differential equations by bookkeeping, c prime equals N times v of c. But chemical rate constants differ by factors of a thousand or a billion: one reaction races ahead and finishes, and an explicit solver pays for the race for the rest of the run. That is stiffness.
  2. Integrated rate laws. First order gives A equals A 0 times e to the minus k t, with a half-life of ln 2 over k that does not depend on A 0. Second order gives one over A equals one over A 0 plus k t, and a half-life that grows as the sample dilutes. Take A 0 equal to 2 and k equal to 0.5. At time 4, first order has left 0.2707, with half-life 1.386. Second order has left 0.4, with half-life exactly 1.
  3. Reversible reactions. A converts to B at rate k f and back at rate k r. A relaxes to A equilibrium equals k r over k f plus k r times A 0, at the rate k f plus k r, the nonzero eigenvalue. With k f equal 2 and k r equal 1, K is 2, A equilibrium is one third, and the rate is 3. A at time 3 is 0.3334, against the closed form 0.33342.
  4. Now a chain, A to B to C. The intermediate B rises, peaks, and falls. For k 1 equal 1 and k 2 equal 0.25 the peak is at 1.848 with height 0.6300, and C at time 8 is 0.8197. If k 2 is much larger than k 1, B is consumed as fast as it is made, so set B prime to zero: B is about k 1 over k 2 times A. This is the quasi-steady-state approximation. For k 2 equal to 20 the exact ratio is 1 over 19, or 0.0526, against 0.05.
  5. Enzymes. An enzyme E binds a substrate S, forms a complex C, and releases a product P. Binding is fast and the enzyme is scarce, so set C prime to zero. The product rate is V max times S over K m plus S, saturating at V max. It holds when epsilon, E 0 over K m plus S 0, is small. With k 1 equal 10, k minus 1 equal 5, k 2 equal 1, K m is 0.6 and S 0 is 1. The largest gap is 0.22 percent of S 0 at epsilon 0.00625, 2.2 percent at 0.0625, and 10.7 percent at 0.3125.
  6. Can chemistry oscillate? Hold two reservoirs constant and the Brusselator can. Its fixed point is a, b over a, with trace b minus 1 minus a squared and determinant a squared. For a equal 1 the trace changes sign at b equal 2: a Hopf bifurcation. At b equal 1.8 the eigenvalues are minus 0.1 plus or minus 0.995 i, a stable spiral. At b equal 3 they are 0.5 plus or minus 0.866 i, and every start winds out to a closed curve, with period 7.157 and x from 0.3706 to 3.7518.
  7. Now stiffness, measured. Take the chain with k 1 equal 1 and k 2 equal 1000. By time 0.01 the fast mode is 4.5 times 10 to the minus 5: invisible. Yet RK4 forces h below 2.785 times 10 to the minus 3, whatever accuracy requires. At h equal 0.0027 RK4 takes 1852 steps and is accurate to 10 to the minus 13. At 0.0028 it blows up at step 644. Backward Euler at h equal 0.1 needs 50 steps, 37 times fewer.
  8. Robertson's problem is the standard stiff chemical test, with rates 0.04, 10 to the 4, and 3 times 10 to the 7. The Jacobian has an eigenvalue of minus 3392.8, so RK4 needs h below 8.2 times 10 to the minus 4. It is stable at 8 times 10 to the minus 4 with 50000 steps, and blows up near t equal 31 at 10 to the minus 3. The trapezoid rule is A-stable, yet at h equal 1 it returns a negative concentration: it rings instead of dying.
  9. Everything so far assumed a well-stirred flask. In a gel or a tissue the species also diffuse, and the Jacobian becomes J minus q squared times the diagonal of D x and D y. A mode grows when b exceeds 1 plus a times the square root of D x over D y, all squared. With a equal 1, D x equal 1 and D y equal 10, the threshold is 1.7325, below the Hopf value 2. At b equal 1.9 the reaction alone is stable, yet q squared equal 0.4 has determinant minus 0.6 and grows. That is a Turing instability, the subject of the next lesson, on reaction diffusion.

Mass action: networks as ODEs

An elementary reaction such as A+B→CA + B \to C with rate constant kk proceeds at the rate v=k [A][B]v = k\,[A][B]: the product of the reactant concentrations, one factor per molecule that takes part. A network becomes a system of ODEs by bookkeeping: reaction jj has rate vj(c)v_j(\mathbf c) and a stoichiometric column (the net change in each species per event), so

dcdt=N v(c).\frac{d\mathbf c}{dt} = N\,\mathbf v(\mathbf c).

For A+B→CA + B \to C: A′=B′=−kABA' = B' = -kAB and C′=+kABC' = +kAB. SIR is the same recipe: S+I→2IS + I \to 2I at rate βSI\beta S I and I→RI \to R at rate γI\gamma I.

Two things follow without solving. Conservation laws: any m\mathbf m with mTN=0\mathbf m^{T}N = 0 makes m⋅c\mathbf m \cdot \mathbf c constant; here A+CA + C and B+CB + C, as S+I+RS + I + R in SIR. And non-negativity: every rate vanishes when its reactant is absent.

Orders and integrated rate laws. A reaction is of order nn in AA if A′=−kAnA' = -kA^n. First order separates to A=A0e−ktA = A_0e^{-kt} with half-life ln⁡2/k\ln 2/k, independent of A0A_0: radioactive decay. Second order, A′=−kA2A' = -kA^2, gives 1/A=1/A0+kt1/A = 1/A_0 + kt and half-life 1/(kA0)1/(kA_0), which grows as the sample dilutes. Take A0=2A_0 = 2 and k=0.5k = 0.5. First order gives A(4)=2e−2=0.2707A(4) = 2e^{-2} = 0.2707 and a half-life of 1.3861.386. Second order gives A(4)=1/(0.5+2)=0.4A(4) = 1/(0.5 + 2) = 0.4 and a half-life of exactly 11. The order is measured, by how the half-life depends on A0A_0; it equals the molecularity only for an elementary step.

◆ Interactive Mass-action rate laws: first against second order, a reversible reaction, and the chain A to B to C with its quasi-steady line

Loading simulation…

Reversible reactions and equilibrium

Let A⇌BA \rightleftharpoons B with forward constant kfk_f and reverse constant krk_r. Then A′=−kfA+krBA' = -k_fA + k_rB, and with A+B=A0A + B = A_0 this is one linear equation: A′=krA0−(kf+kr)AA' = k_rA_0 - (k_f + k_r)A. The solution relaxes to the equilibrium

Aeq=krkf+krA0,K=BeqAeq=kfkr,A(t)=Aeq+(A0−Aeq)e−(kf+kr)t.A_{eq} = \frac{k_r}{k_f + k_r}A_0, \qquad K = \frac{B_{eq}}{A_{eq}} = \frac{k_f}{k_r}, \qquad A(t) = A_{eq} + (A_0 - A_{eq})e^{-(k_f + k_r)t}.

The rate is the sum kf+krk_f + k_r, the nonzero eigenvalue of the system. With kf=2k_f = 2 and kr=1k_r = 1, A0=1A_0 = 1: K=2K = 2, Aeq=1/3A_{eq} = 1/3, the relaxation rate is 33, and A(3)=0.3334A(3) = 0.3334 against the closed form 0.333420.33342 (sim and scipy). At equilibrium each reaction is balanced by its reverse (detailed balance), which is why KK is a ratio of rate constants. The relaxation rates of any first-order network are eigenvalues of its rate matrix: 7.2 Linear Systems & the Eigenvalue Method.

Consecutive reactions and the quasi-steady state

Now A→k1B→k2CA \xrightarrow{k_1} B \xrightarrow{k_2} C from A=1A = 1. Then A=e−k1tA = e^{-k_1t} and B′=k1e−k1t−k2BB' = k_1e^{-k_1t} - k_2B, a linear equation with solution

B(t)=k1k2−k1(e−k1t−e−k2t),C=1−A−B.B(t) = \frac{k_1}{k_2 - k_1}\left(e^{-k_1t} - e^{-k_2t}\right), \qquad C = 1 - A - B.

The intermediate rises, peaks where B′=0B' = 0, that is at t∗=ln⁡(k2/k1)/(k2−k1)t^* = \ln(k_2/k_1)/(k_2 - k_1), and falls. For k1=1k_1 = 1, k2=0.25k_2 = 0.25: t∗=1.848t^* = 1.848 and Bmax=0.6300B_{max} = 0.6300, and C(8)=0.8197C(8) = 0.8197 (closed form, sim and scipy agree).

The quasi-steady state. Suppose instead k2≫k1k_2 \gg k_1: BB is consumed as fast as it is made, so it stays small and nearly constant relative to AA. Setting B′≈0B' \approx 0 gives

B≈k1k2A.B \approx \frac{k_1}{k_2}A.

This is the quasi-steady-state approximation (QSSA): a fast variable is slaved to the slow ones. For k1=1k_1 = 1 and k2=20k_2 = 20 the exact ratio after the first transient is B/A=k1/(k2−k1)=1/19=0.0526B/A = k_1/(k_2 - k_1) = 1/19 = 0.0526, against the QSSA’s 0.050.05: a relative error of k1/k2k_1/k_2. The transient itself is short: BB peaks at t=0.158t = 0.158 with the tiny value 0.04270.0427. The error is small exactly when the rates are far apart, which is also when the problem is stiff.

Lindemann’s mechanism is the classical use. A molecule gains energy by collision with MM and then reacts: A+M⇌A∗+MA + M \rightleftharpoons A^* + M (constants k1k_1, k−1k_{-1}), A∗→PA^* \to P (k2k_2). Setting (A∗)′=0(A^*)' = 0 gives A∗=k1MA/(k−1M+k2)A^* = k_1MA/(k_{-1}M + k_2) and an effective rate

dPdt=keffA,keff=k1k2Mk−1M+k2.\frac{dP}{dt} = k_{\text{eff}}A, \qquad k_{\text{eff}} = \frac{k_1k_2M}{k_{-1}M + k_2}.

At low MM this is k1Mk_1M, second order (collisions limit). At high MM it is k1k2/k−1k_1k_2/k_{-1}, first order. The crossover is at M=k2/k−1M = k_2/k_{-1}. Take k1=0.1k_1 = 0.1, k−1=10k_{-1} = 10, k2=1k_2 = 1 and hold MM fixed, so the system is linear. Its slow eigenvalue (the exact decay rate of A+A∗A + A^*) against keffk_{\text{eff}}, with the fast one:

MMexact slow ratekeffk_{\text{eff}}fast rate
0.010.019.0902×10−49.0902\times10^{-4}9.0909×10−49.0909\times10^{-4}1.101.10
0.10.14.9875×10−34.9875\times10^{-3}5.0000×10−35.0000\times10^{-3}2.012.01
119.0163×10−39.0163\times10^{-3}9.0909×10−39.0909\times10^{-3}11.0911.09
10109.8049×10−39.8049\times10^{-3}9.9010×10−39.9010\times10^{-3}101.99101.99
1001009.8913×10−39.8913\times10^{-3}9.9900×10−39.9900\times10^{-3}1011.01011.0

(Eigenvalues from the sim’s 2×22\times2 solver and numpy; the last column is the magnitude of the fast eigenvalue.) The relative error settles near 1%1\%, which is k1/k−1k_1/k_{-1}: the QSSA ignores the small reservoir held in A∗A^*.

Michaelis-Menten kinetics

An enzyme EE binds a substrate SS, forms a complex CC and releases product PP: E+S⇌C→E+PE + S \rightleftharpoons C \to E + P, with constants k1k_1, k−1k_{-1}, k2k_2. Mass action, with the enzyme conserved (E=E0−CE = E_0 - C), gives

S′=−k1(E0−C)S+k−1C,C′=k1(E0−C)S−(k−1+k2)C,P′=k2C.S' = -k_1(E_0 - C)S + k_{-1}C, \quad C' = k_1(E_0 - C)S - (k_{-1} + k_2)C, \quad P' = k_2C.

Binding is fast and the enzyme is scarce, so CC settles quickly to a slowly drifting value. Set C′=0C' = 0 (the Briggs-Haldane form of the QSSA):

C=E0SKm+S,Km=k−1+k2k1,dPdt=VmaxSKm+S,Vmax=k2E0.C = \frac{E_0S}{K_m + S}, \qquad K_m = \frac{k_{-1} + k_2}{k_1}, \qquad \frac{dP}{dt} = V_{max}\frac{S}{K_m + S}, \quad V_{max} = k_2E_0.

This is the Michaelis-Menten law: rate proportional to SS when S≪KmS \ll K_m (first order), saturating at VmaxV_{max} when S≫KmS \gg K_m (zero order, the enzyme is the bottleneck).

When is the QSSA valid? The complex forms on the time 1/(k1(Km+S0))1/(k_1(K_m + S_0)) and the substrate depletes on the much longer time (Km+S0)/Vmax(K_m + S_0)/V_{max}. A sufficient condition is that the enzyme is a small fraction of what is available,

ε=E0Km+S0≪1\varepsilon = \frac{E_0}{K_m + S_0} \ll 1

(Segel and Slemrod’s condition, from memory). The reduced model P′=Vmax(S0−P)/(Km+S0−P)P' = V_{max}(S_0 - P)/(K_m + S_0 - P) treats the substrate locked in CC as zero, so it errs by about C≤E0C \le E_0. Test it with k1=10k_1 = 10, k−1=5k_{-1} = 5, k2=1k_2 = 1 (Km=0.6K_m = 0.6) and S0=1S_0 = 1, and compare the full ODE with the reduced equation:

E0E_0ε\varepsilonlargest gap between the two products% of S0S_0at time
0.010.010.006250.006250.002220.002220.220.22175175
0.10.10.06250.06250.022080.022082.22.217.717.7
0.50.50.31250.31250.10660.106610.710.73.673.67

(Sim and scipy Radau at tolerance 10−1110^{-11} agree to four digits.) The error grows roughly in proportion to ε\varepsilon, and at ε=0.31\varepsilon = 0.31 the reduced curve is visibly wrong.

◆ Interactive Michaelis-Menten: full mass action against the quasi-steady reduction; raise the enzyme concentration and the reduction fails

Loading simulation…

Oscillating reactions: the Brusselator

Can a chemical system oscillate? Not for the closed reactions above: they relax monotonically to equilibrium or are driven by constant reservoirs, as in a living cell. Hold the reservoirs AA and BB constant in the Brusselator scheme: A→XA \to X, then 2X+Y→3X2X + Y \to 3X (autocatalysis), B+X→Y+DB + X \to Y + D, and X→EX \to E. Mass action gives, with rate constants set to 11,

x′=a−(b+1)x+x2y,y′=bx−x2y.x' = a - (b + 1)x + x^2y, \qquad y' = bx - x^2y.

The only fixed point is (x∗,y∗)=(a,b/a)(x^*, y^*) = (a, b/a). The Jacobian there is (b−1a2−b−a2)\begin{pmatrix} b - 1 & a^2 \\ -b & -a^2\end{pmatrix}, so

τ=b−1−a2,Δ=a2>0.\tau = b - 1 - a^2, \qquad \Delta = a^2 \gt 0.

The determinant is positive, so the point is a node or spiral, stable iff τ<0\tau \lt 0. As bb passes 1+a21 + a^2 the trace changes sign and the eigenvalues 12[τ±τ2−4a2]\tfrac12\left[\tau \pm \sqrt{\tau^2 - 4a^2}\right] cross the imaginary axis at ±ia\pm ia: a Hopf bifurcation, exactly the mechanism of 7.9 Limit Cycles & Hopf Bifurcations, with period near 2π/a2\pi/a at onset. For a=1a = 1 the threshold is b=2b = 2. At b=1.8b = 1.8 the eigenvalues are −0.1±0.995i-0.1 \pm 0.995i, a stable spiral. At b=3b = 3 they are 0.5±0.866i0.5 \pm 0.866i and every start winds out to a closed curve (the cycle’s existence follows from the Poincare-Bendixson theorem, from memory).

aabbperiodrange of xx
112.12.16.3226.3220.70950.7095 to 1.47311.4731
112.22.26.3706.3700.62290.6229 to 1.75131.7513
11337.1577.1570.37060.3706 to 3.75183.7518
1.51.5556.0966.0960.36310.3631 to 6.54416.5441

(Sim and scipy agree on periods to 10−610^{-6} and on ranges to four digits.) The period 6.3226.322 just above onset is close to 2π=6.2832\pi = 6.283 and drifts up as the cycle grows; for a=1.5a = 1.5, b=5b = 5 it is far from 2π/a=4.192\pi/a = 4.19. The amplitude grows like b−bc\sqrt{b - b_c} for a supercritical Hopf: half-ranges of xx are 0.3820.382 and 0.5640.564 at b=2.1b = 2.1 and 2.22.2, a ratio of 1.481.48 against 2=1.41\sqrt2 = 1.41.

◆ Interactive The Brusselator: slide b through the Hopf threshold 1 + a^2, click to add starts, and read the eigenvalues and the period

Loading simulation…

The real Belousov-Zhabotinsky reaction (bromate, malonic acid and a metal catalyst, cycling through colours) is modelled by the three-variable Oregonator (from memory, not run here): autocatalysis plus delayed negative feedback again, and widely separated timescales that make it stiff.

Stiffness, measured

Take the chain with a fast second step: k1=1k_1 = 1 and k2=1000k_2 = 1000, from (1,0,0)(1, 0, 0) to T=5T = 5. The Jacobian has eigenvalues −1-1, −1000-1000 and 00. By t=0.01t = 0.01 the fast mode e−1000te^{-1000t} is 4.5×10−54.5\times10^{-5}: invisible, and the solution is smooth, B≈A/1000B \approx A/1000. Yet recall from 4.5 Stability Regions & Stiffness that RK4 on y′=λyy' = \lambda y is stable iff ∣R(hλ)∣≤1|R(h\lambda)| \le 1 with R(z)=1+z+z2/2+z3/6+z4/24R(z) = 1 + z + z^2/2 + z^3/6 + z^4/24, which holds for hλ≥−2.7853h\lambda \ge -2.7853 (the root of ∣R∣=1|R| = 1, recomputed here). The fast mode forces

h<2.78531000=2.785×10−3,h \lt \frac{2.7853}{1000} = 2.785\times10^{-3},

whatever accuracy requires. The measurements:

methodstep hhstepsmax error at t=5t = 5
RK40.00270.002718521852<10−13\lt 10^{-13}
RK40.00280.0028exceeds 10001000 at step 644644∞\infty
backward Euler0.50.510101.06×10−21.06\times10^{-2}
backward Euler0.10.150501.78×10−31.78\times10^{-3}
backward Euler0.010.015005001.70×10−41.70\times10^{-4}
trapezoid0.50.510101.62×10−31.62\times10^{-3}
trapezoid0.10.150501.64×10−41.64\times10^{-4}
trapezoid0.050.051001007.4×10−67.4\times10^{-6}

(Errors against the exact solution; sim and numpy agree to four digits.) At h=0.0028h = 0.0028 the per-step factor is ∣R(−2.8)∣=1.0224|R(-2.8)| = 1.0224, and the fast mode grows by 1.02241786≈10171.0224^{1786} \approx 10^{17} over the run. RK4 at h=0.0027h = 0.0027 is accurate to 10−1310^{-13}: it pays 18521852 steps for accuracy nobody asked for, while backward Euler at h=0.1h = 0.1 meets a 2×10−32\times10^{-3} tolerance with 5050 steps, 3737 times fewer. Steps are not cost: an implicit step solves a linear system (a Newton iteration if the right side is nonlinear). But the explicit limit scales as 1/k21/k_2 and the implicit step does not care: at k2=106k_2 = 10^6 the RK4 count grows a thousandfold and the implicit count stays put.

Robertson’s problem, the standard stiff chemical test of 4.5 Stability Regions & Stiffness, uses rates 0.040.04, 10410^4 and 3×1073\times10^7. The sim reruns it with fixed steps on [0,40][0, 40] from (1,0,0)(1, 0, 0). Radau and BDF at tolerance 10−1210^{-12} give y(40)=(0.715827,9.1855×10−6,0.284164)y(40) = (0.715827, 9.1855\times10^{-6}, 0.284164), and the Jacobian there has an eigenvalue −3392.8-3392.8, so RK4 needs h<2.7853/3392.8=8.2×10−4h \lt 2.7853/3392.8 = 8.2\times10^{-4}.

methodstep hhstepsy1(40)y_1(40) and its error
RK48×10−48\times10^{-4}50000500000.7158270.715827 (error 9×10−129\times10^{-12})
RK410−310^{-3}blows up at t≈31t \approx 31none
backward Euler1140400.719190.71919 (+3.4×10−3+3.4\times10^{-3})
backward Euler0.10.14004000.716170.71617 (+3.5×10−4+3.5\times10^{-4})
trapezoid1140400.63160.6316 (−0.084-0.084, with y2<0y_2 \lt 0)
trapezoid0.10.14004000.714590.71459 (−1.2×10−3-1.2\times10^{-3})

The RK4 row sits right on the predicted limit: stable at 8×10−48\times10^{-4}, dead at 10−310^{-3}. The trapezoid rule is A-stable but at h=1h = 1 it returns a negative concentration, and at h=0.1h = 0.1 it is three and a half times worse than backward Euler. The reason is the amplification factor at large ∣z∣|z|: Rtrap→−1R_{\text{trap}} \to -1, so the fast transient rings instead of dying, while RBE→0R_{BE} \to 0 (L-stability). This is the stable-is-not-accurate warning of the stiffness lesson, now on chemistry; production codes use BDF and Radau.

◆ Interactive Stiffness: choose RK4, backward Euler or the trapezoid rule, slide the step, and watch the explicit run explode just above 2.785 over the fast rate

Loading simulation…

Where this goes next

Everything above assumed a well-stirred flask. In a gel or a tissue the species also diffuse, and the Brusselator becomes ∂tx=Dx∇2x+…\partial_tx = D_x\nabla^2x + \dots, ∂ty=Dy∇2y+…\partial_ty = D_y\nabla^2y + \dots. Diffusion is stabilising for a single species, yet it can destabilise the pair. Perturb about (a,b/a)(a, b/a) with eiqxe^{iqx}: the Jacobian becomes J−q2 diag(Dx,Dy)J - q^2\,\text{diag}(D_x, D_y), with determinant a2+(a2Dx−(b−1)Dy)q2+DxDyq4a^2 + \left(a^2D_x - (b - 1)D_y\right)q^2 + D_xD_yq^4. This can go negative, and a mode grows, iff

b>(1+aDx/Dy)2,q2=aDxDy at threshold.b \gt \left(1 + a\sqrt{D_x/D_y}\right)^2, \qquad q^2 = \frac{a}{\sqrt{D_xD_y}} \text{ at threshold.}

With a=1a = 1, Dx=1D_x = 1, Dy=10D_y = 10 the threshold is b=1.7325b = 1.7325, below the Hopf value 22. At b=1.9b = 1.9 the reaction alone is a stable spiral (trace −0.1-0.1), but the mode q2=0.4q^2 = 0.4 has determinant −0.6-0.6 and grows: a Turing instability, where a stable chemistry plus a faster-diffusing inhibitor makes stationary stripes and spots. That is the subject of 15.5 Reaction–Diffusion & Turing Patterns, together with Gray-Scott patterns and the travelling fronts of Fisher-KPP. One honest limit: this only shows where the uniform state loses stability; the pattern itself needs a nonlinear simulation, built next lesson.

The exercise: Michaelis-Menten, Lindemann, backward Euler and the Brusselator

First the playground: the stiff chain with RK4 and backward Euler written as matrix powers, so the stability threshold is visible.

🐍 PythonStiff chain: RK4 against backward Euler
⌘/Ctrl + Enter
🧪 Coding exerciseExercise: rate laws, backward Euler, Brusselator eigenvalues, the chain peak and the RK4 limit
⌘/Ctrl + Enter

Concept checks

✅ Concept check

For a second-order reaction $A' = -kA^2$, what happens to the half-life as the reaction proceeds?

✅ Concept check

For $A \rightleftharpoons B$ with $k_f = 2$ and $k_r = 1$, at what rate does the system relax to equilibrium?

✅ Concept check

Which condition makes the Michaelis-Menten (quasi-steady) reduction accurate?

✅ Concept check

For the Brusselator with $a = 2$, at what value of $b$ does the fixed point lose stability?

✅ Concept checkselect all that apply

Select all true statements about the chain $A \to B \to C$ with $k_1 = 1$ and $k_2 = 1000$.

Problems

✏️ Pen & paperHalf-life and time to 99 percent●○○

A first-order reaction has a half-life of 2020 s. Find kk and the time for 99%99\% of the reactant to be consumed.

Show worked solution

k=ln⁡2/20=0.034657 s−1k = \ln 2/20 = 0.034657\ \text{s}^{-1}. For 99%99\% consumed, e−kt=0.01e^{-kt} = 0.01, so t=ln⁡100/k=4.6052/0.034657=132.9t = \ln 100/k = 4.6052/0.034657 = 132.9 s (about 6.646.64 half-lives, as 26.64=1002^{6.64} = 100).

✏️ Pen & paperReversible reaction: equilibrium and relaxation●●○

For A⇌BA \rightleftharpoons B with kf=3k_f = 3 and kr=1k_r = 1 and A(0)=1A(0) = 1, B(0)=0B(0) = 0: find KK, AeqA_{eq} and A(t)A(t), and the time for AA to come within 0.010.01 of its equilibrium.

Show worked solution

K=3K = 3, Aeq=kr/(kf+kr)=1/4A_{eq} = k_r/(k_f + k_r) = 1/4, and A(t)=14+34e−4tA(t) = \tfrac14 + \tfrac34e^{-4t}. Setting 34e−4t=0.01\tfrac34e^{-4t} = 0.01 gives e−4t=1/75e^{-4t} = 1/75, so t=ln⁡75/4=1.0794t = \ln 75/4 = 1.0794.

✏️ Pen & paperMichaelis-Menten data●●○

An enzyme has Vmax=2V_{max} = 2 and Km=3K_m = 3 (in consistent units). Find the rate at S=1S = 1, the substrate level where the rate is 1.51.5, and the substrate level where the rate is half of VmaxV_{max}. What does the answer say about KmK_m?

Show worked solution

v(1)=2⋅1/(3+1)=0.5v(1) = 2 \cdot 1/(3 + 1) = 0.5. For v=1.5v = 1.5: 2S=1.5(3+S)2S = 1.5(3 + S), so 0.5S=4.50.5S = 4.5 and S=9S = 9. For v=Vmax/2=1v = V_{max}/2 = 1: 2S=3+S2S = 3 + S gives S=3=KmS = 3 = K_m. So KmK_m is the half-saturation substrate level.

✏️ Pen & paperBrusselator at a different feed, and the step budget●●●

(a) For a=2a = 2, find the Hopf value of bb and the oscillation period at onset. (b) For the chain with k1=1k_1 = 1 and k2=104k_2 = 10^4 run to T=10T = 10, how many RK4 steps does stability force, and how many does backward Euler need at h=0.1h = 0.1?

Show worked solution

(a) bc=1+a2=5b_c = 1 + a^2 = 5, the determinant is a2=4a^2 = 4 and at onset the eigenvalues are ±2i\pm 2i, so the period is 2π/2=π=3.14162\pi/2 = \pi = 3.1416. (b) Stability needs h<2.7853/104=2.7853×10−4h \lt 2.7853/10^4 = 2.7853\times10^{-4}, so n>10/(2.7853×10−4)=35903.5n \gt 10/(2.7853\times10^{-4}) = 35903.5: at least 3590435904 steps, or 143616143616 evaluations of the right side at four per step. Backward Euler at h=0.1h = 0.1 takes 100100 steps, whatever k2k_2 is.

Key Takeaways

  • Mass action: rate =k×= k \times the product of reactant concentrations, c′=Nv(c)\mathbf c' = N\mathbf v(\mathbf c); SIR is a reaction network.
  • First-order half-lives are constant, second-order ones grow; a reversible reaction relaxes at kf+krk_f + k_r to K=kf/krK = k_f/k_r.
  • For A→B→CA \to B \to C, BB peaks at t∗=ln⁡(k2/k1)/(k2−k1)t^* = \ln(k_2/k_1)/(k_2 - k_1); the quasi-steady approximation B≈(k1/k2)AB \approx (k_1/k_2)A has relative error about k1/k2k_1/k_2.
  • Michaelis-Menten, v=VmaxS/(Km+S)v = V_{max}S/(K_m + S), follows from C′≈0C' \approx 0 and holds when ε\varepsilon is small (0.22%0.22\% error at 0.0060.006, 10.7%10.7\% at 0.310.31).
  • The Brusselator has fixed point (a,b/a)(a, b/a), trace b−1−a2b - 1 - a^2, determinant a2a^2: a Hopf bifurcation at b=1+a2b = 1 + a^2, with period near 2π/a2\pi/a at onset (6.326.32 at b=2.1b = 2.1, a=1a = 1).
  • Stiffness is a step set by stability, not accuracy: RK4 on a rate of 10001000 needs h<2.785×10−3h \lt 2.785\times10^{-3} (18521852 steps) while backward Euler takes 5050 steps at h=0.1h = 0.1 for 2×10−32\times10^{-3} accuracy.
  • A-stability is not enough: on Robertson the trapezoid rule at h=1h = 1 gives a negative concentration; backward Euler is 0.5%0.5\% off.
  • Add diffusion to the Brusselator and the stable state can become unstable to patterned modes when b>(1+aDx/Dy)2b \gt (1 + a\sqrt{D_x/D_y})^2: the Turing instability of 15.5 Reaction–Diffusion & Turing Patterns.
Finished this lesson?

Mark it complete to track your progress (saved in this browser). Reading every section does this for you.