💡 The Idea, Step by Step
Look at a plastic shopping bag, a foam coffee cup, a length of PVC pipe. Each one is essentially a single molecule built from millions of identical little pieces snapped end-to-end into one impossibly long chain. The little piece is the monomer; the giant chain is the polymer. Free-radical polymerization is one recipe for building those chains, and it behaves like a conga line: one dancer grabs the next, who instantly grabs the next, and the whole line grows from a single fast-moving end.
That fast-moving end is a radical — an atom carrying one unpaired electron, desperate to pair up. You make the first radicals by gently warming an initiator such as AIBN, which splits in two: $I \rightarrow 2R\bullet$. Each radical pounces on a monomer's double bond and becomes a slightly longer radical, then does it again, and again — this repeated step is propagation, $P_n\bullet + M \rightarrow P_{n+1}\bullet$. The chain lengthens at the rate $R_p = k_p [M][P\bullet]$. Put in real styrene numbers ($k_p = 340$, $[M] = 8.7$ M, $[P\bullet] \approx 2.7\times10^{-8}$ M) and you get about $8\times10^{-5}$ moles of monomer consumed per litre every second — a single chain can swallow roughly a thousand monomers in about a second before it dies.
Why doesn't the radical population just keep climbing? Because whenever two radical chain-ends meet they snuff each other out (termination, $2P\bullet \rightarrow$ dead polymer). Radicals are born in pairs and die in pairs, so they settle into a steady state where birth rate equals death rate: $2 f k_d [I] = 2 k_t [P\bullet]^2$, which solves to $[P\bullet] = \sqrt{f k_d [I]/k_t}$. Substitute that back and the rate becomes $R_p = k_p [M]\sqrt{f k_d [I]/k_t}$ — notice it depends on the square root of initiator, so doubling $[I]$ speeds things up only $1.41\times$. The sliders map straight onto these symbols: $[M]$ and $[I]$ are the concentrations, $k_p$ and $k_t$ the propagation and termination constants, $f$ the fraction of radicals that actually start a chain, and temperature $T$ feeds the Arrhenius decomposition rate $k_d$.
Try this in the sim above. First, push $[I]$ from low to high and watch $R_p$ creep up only as $\sqrt{[I]}$ while the average chain length DPₙ shrinks — more chain-ends means more termination. Next, raise the temperature and notice Mₙ falls even though the reaction runs faster: the initiator falls apart faster than propagation can use the extra radicals. Finally, switch the preset from Styrene to Ethylene (LDPE, high pressure) and compare the molecular weights the two recipes reach.
📐 Kinetics & Mayo Equation
The Four Elementary Steps
$$\text{Initiation: } I \xrightarrow{k_d} 2R\bullet, \quad R\bullet + M \xrightarrow{k_i} P_1\bullet$$
$$\text{Propagation: } P_n\bullet + M \xrightarrow{k_p} P_{n+1}\bullet$$
$$\text{Chain Transfer: } P_n\bullet + S \xrightarrow{k_{tr}} P_n + S\bullet$$
$$\text{Termination: } P_n\bullet + P_m\bullet \xrightarrow{k_t} P_{n+m} \text{ (combination) or } P_n + P_m \text{ (disproportionation)}$$
Steady-State Rate of Polymerization
$$R_p = -\frac{d[M]}{dt} = k_p [M] \left(\frac{f \cdot k_d \cdot [I]}{k_t}\right)^{1/2}$$
The square-root dependence on [I] is the signature of free-radical polymerization. Doubling [I] increases rate only by 1.41×, not 2×.
Symbol Definitions
| Symbol | Meaning | Unit |
|---|
| $k_d$ | Initiator dissociation rate constant | s⁻¹ |
| $k_p$ | Propagation rate constant | L mol⁻¹ s⁻¹ |
| $k_t$ | Termination rate constant (sum of comb + disp) | L mol⁻¹ s⁻¹ |
| $k_{tr}$ | Chain transfer rate constant | L mol⁻¹ s⁻¹ |
| $f$ | Initiator efficiency (fraction of R• that initiate) | 0 ≤ f ≤ 1 |
| $DP_n$ | Number-average degree of polymerization | — |
| $\bar{M}_n, \bar{M}_w$ | Number-, weight-average molecular weights | g/mol |
| PDI | Polydispersity index = Mw/Mₙ | ≥ 1 |
Step-by-Step Derivation: Mayo Equation
1Initiation rate (assuming f for efficiency): Initiator I cleaves at rate $k_d[I]$ giving 2 radicals; only fraction $f$ start chains. Rate of radical generation = $R_i = 2 f k_d [I]$.
2Steady-state hypothesis: The total radical concentration $[P\bullet]$ becomes constant after a brief induction period. Termination is bimolecular: $R_t = 2 k_t [P\bullet]^2$. At steady state: $R_i = R_t$, so $[P\bullet] = (f k_d [I] / k_t)^{1/2}$.
3Propagation rate: $R_p = k_p [M] [P\bullet] = k_p [M] (f k_d [I] / k_t)^{1/2}$. The √[I] dependence is observable experimentally — confirms the radical mechanism.
4Kinetic chain length ν: Average number of monomers added per radical before termination: $\nu = R_p / R_i = k_p [M] / (2 k_t [P\bullet]) = k_p [M] / (2 (f k_d k_t [I])^{1/2})$. Higher [M], lower [I], smaller kₜ → higher ν.
5Degree of polymerization DPₙ depends on termination mode:
(a) Pure combination (e.g., styrene): $DP_n = 2\nu$
(b) Pure disproportionation (e.g., MMA at high T): $DP_n = \nu$
(c) With chain transfer to solvent S: Mayo equation $\frac{1}{DP_n} = \frac{1}{DP_{n,0}} + C_S \frac{[S]}{[M]}$ where $C_S = k_{tr,S}/k_p$.
6Why PDI ≈ 1.5–2.0 is theoretical limit? In ideal free-radical polymerization, the random termination gives a Schulz-Flory distribution: PDI = 1.5 (combination) or 2.0 (disproportionation). Experimental PDIs of 1.8–2.5 are typical. Living polymerization (ATRP, RAFT) gives PDI < 1.2 because all chains start at the same time and terminate together.
Worked Example — Styrene Polymerization
Conditions: [Styrene] = 8.7 M (bulk), [AIBN] = 0.01 M, T = 60°C. Constants: $k_d = 8.5 \times 10^{-6}$ s⁻¹, $k_p = 340$ L/mol·s, $k_t = 7.2 \times 10^7$ L/mol·s, f = 0.6.
Calculate Rₚ:
$[P\bullet] = ((0.6)(8.5 \times 10^{-6})(0.01)/(7.2 \times 10^7))^{1/2} = (7.08 \times 10^{-16})^{1/2} = 2.7 \times 10^{-8}$ M
$R_p = (340)(8.7)(2.7 \times 10^{-8}) = 7.9 \times 10^{-5}$ M/s
Calculate DPₙ:
$\nu = R_p / R_i = R_p / (2 f k_d [I]) = 7.9 \times 10^{-5} / (2 \times 0.6 \times 8.5 \times 10^{-6} \times 0.01) = 772$
For styrene (combination): $DP_n = 2\nu = 1540$. With chain transfer to monomer ($C_M \approx 6 \times 10^{-5}$): $1/DP_n = 1/1540 + 6 \times 10^{-5}$, giving DPₙ ≈ 1410 — within the experimentally observed range.
$\bar{M}_n = 1410 \times 104$ g/mol ≈ 147 kg/mol
📚 References:
• Odian, G. — Principles of Polymerization, 4th Ed., Wiley (2004), Ch. 3
• Cowie, J.M.G. — Polymers: Chemistry & Physics of Modern Materials, 3rd Ed., CRC (2007), Ch. 6
• Stevens, M.P. — Polymer Chemistry: An Introduction, 3rd Ed., Oxford UP (1999)
• Flory, P.J. — Principles of Polymer Chemistry, Cornell UP (1953) — classic
• Mayo, F.R. — J. Am. Chem. Soc. 65, 2324 (1943) — chain transfer equation
• Atkins & de Paula — Physical Chemistry, 11th Ed., Ch. 17F: "Polymers"
❓ Frequently Asked Questions
🧪 ConceptualWhy is the rate proportional to √[I] and not [I]?▼
Because radicals are created in pairs (initiator splits into 2R•) but destroyed in pairs (P• + P• → polymer). At steady state, the rate of creation equals the rate of destruction: $2fk_d[I] = 2k_t[P\bullet]^2$. Solving for radical concentration gives $[P\bullet] \propto [I]^{1/2}$. Since propagation rate depends linearly on [P•], the overall polymerization rate goes as √[I]. This square-root law is THE fingerprint of radical chain mechanisms — observed experimentally and used historically (in the 1930s–40s) to confirm radical involvement.Key Takeaway: Doubling initiator only multiplies rate by 1.41× because radicals self-destruct in pairs — the √[I] law diagnoses radical mechanism.
🌍 Real LifeWhat everyday products come from free radical polymerization?▼
A massive list: low-density polyethylene (LDPE — plastic bags, squeeze bottles, made at 1500–3000 atm with O₂ initiator); polystyrene (Styrofoam cups, packaging foam, made from styrene + AIBN/peroxide); poly(vinyl chloride) PVC (pipes, vinyl flooring, made via emulsion polymerization with persulfate); poly(methyl methacrylate) PMMA (Plexiglas, contact lenses, hard transparent panels); polyacrylonitrile (acrylic fibers like Orlon, precursor for carbon fiber); poly(vinyl acetate) PVAc (white school glue, water-based paint base); SBR rubber (tires; styrene-butadiene copolymer via emulsion radical polymerization). Worldwide, free-radical polymers represent ~50% of all synthetic polymers produced — over 200 million tonnes/year.Key Takeaway: LDPE, PS, PVC, PMMA — half the world's plastics — are made by free radical polymerization, mostly via similar AIBN/peroxide initiation.
🔬 SimulationWhat does the molecular weight distribution (MWD) graph show?▼
The MWD plot shows the relative number of polymer chains with each chain length (DP). For free radical polymerization, this distribution is the Schulz-Flory ("most probable") distribution. The simulation generates this distribution dynamically: as polymerization proceeds, the histogram fills with chains of various lengths, peaked around the average DPₙ. Two key averages are shown: Mₙ (number-average, where short chains weigh heavily) and Mw (weight-average, where long chains weigh heavily). The ratio Mw/Mₙ = PDI; for ideal Schulz-Flory, PDI = 1.5 (combination) or 2.0 (disproportionation). Real distributions are slightly broader because of imperfections.Key Takeaway: MWD width directly measures kinetic mechanism — narrow distribution (PDI ≈ 1.5-2) confirms ideal radical polymerization; broader means complications.
💡 Non-ObviousWhy does increasing temperature DECREASE molecular weight?▼
Counterintuitive! Increasing T does increase Rₚ (more propagation), but it also increases initiator decomposition rate $k_d$ EVEN MORE (typical activation energies: $E_a^d = 130$ kJ/mol vs $E_a^p = 30$ kJ/mol). More radicals are generated → more termination events → shorter chains. The expression $\nu \propto k_p [M] / (k_d k_t [I])^{1/2}$ shows that $\nu$ depends on $E_a^p - \frac{1}{2}E_a^d - \frac{1}{2}E_a^t$. Since $E_a^d/2 \approx 65$ kJ/mol >> $E_a^p \approx 30$, raising T reduces the kinetic chain length. This is why polymers are made at moderate temperatures (50-100°C) — too hot gives short chains.Key Takeaway: Higher T = faster reaction BUT shorter chains because k_d's high Ea makes initiator radicals proliferate faster than propagation can use them.
🧮 MathematicalHow does chain transfer to solvent reduce molecular weight?▼
The Mayo equation: $\frac{1}{DP_n} = \frac{1}{DP_{n,0}} + C_S \frac{[S]}{[M]}$ where $C_S = k_{tr,S}/k_p$ is the chain transfer constant. If $C_S = 10^{-3}$ and [S]/[M] = 1, then $1/DP_n$ increases by 0.001, so DPₙ drops from say 5000 to 1000. CCl₄ has $C_S \approx 5$ (huge!) — even small amounts cap chain length to a few hundred. Engineers EXPLOIT this: adding "chain transfer agents" (CTAs, a.k.a. molecular weight regulators) like dodecanethiol intentionally limits MW for processability. Plot $1/DP_n$ vs [S]/[M] gives slope = $C_S$ — standard way to measure transfer constants.Key Takeaway: Mayo equation links chain transfer rate to MW — used both to diagnose unintended transfer and to deliberately control MW with CTAs.
🌌 Deep / AdvancedWhat is the Trommsdorff (gel) effect?▼
As a free-radical polymerization proceeds, viscosity rises dramatically. At ~30-50% conversion, polymer chains become so entangled that diffusion of growing radicals slows enormously. Termination (bimolecular, requires two radicals to find each other) slows down MUCH more than propagation (small monomers still diffuse). Result: $k_t$ drops 100-1000× while $k_p$ stays nearly constant. Ratio $k_p / k_t^{1/2}$ JUMPS, so Rₚ AUTO-ACCELERATES (despite shrinking [M]) and DPₙ shoots up. This "gel effect" or "Norrish-Trommsdorff effect" is named after F.R. Mayo's collaborator. Industrial impact: bulk polymerization can run away thermally if not controlled — many fires in MMA plants. The fix: emulsion polymerization (radicals trapped in tiny droplets, no encounter possible) or temperature/[I] modulation.Key Takeaway: Trommsdorff effect — viscous medium hinders radical-radical termination but not propagation, causing dangerous self-acceleration in bulk radical polymerizations.
🌍 Real LifeWhat's the difference between free radical and "living" polymerization?▼
In conventional radical polymerization, chains have lifetimes of ~1 second — they grow rapidly then die irreversibly through termination. Result: broad MWD (PDI 2+), no control over chain ends, no block copolymers possible. "Living" or controlled radical polymerization (CRP) — including ATRP (Matyjaszewski 1995), RAFT (Rizzardo 1998), and NMP — uses reversible deactivation: a "dormant" form (e.g., ATRP halide-capped chain) interconverts with active radical, suppressing termination. Most chains stay alive throughout. PDI drops to 1.05-1.2; block copolymers become accessible (drug delivery, electronic materials). The 2005 Nobel for olefin metathesis touched related polymer ideas; ATRP/RAFT are major academic and industrial techniques today.Key Takeaway: Living radical polymerization (ATRP/RAFT) suppresses termination via reversible deactivation, giving narrow MW (PDI ~1.1) and enabling block copolymers.
📚 Best Resources for Beginners:
• Odian, G. — Principles of Polymerization, 4th Ed., Wiley (2004) — gold-standard textbook
• LibreTexts Chemistry — Polymer Chemistry: "Chain-growth polymerization"
• Royal Society of Chemistry — Polymer Chemistry journal & ChemEd Resources
• Matyjaszewski Polymer Group at CMU — open lecture notes on CRP/ATRP
⚠️ Common Misconceptions
❌ "All polymer chains formed in radical polymerization have the same length."
✅ Far from it — chains have a distribution of lengths (the Schulz-Flory distribution). Some chains terminate early (short), some grow long; the average is DPₙ but the spread is wide. Polydispersity index (PDI = Mw/Mₙ) is 1.5-2.5 typically. To get uniform chains you need a "living" polymerization technique (ATRP, RAFT, anionic), where all chains start at once and grow synchronously, giving PDI ~1.05.
📖 Reference: Odian — Principles of Polymerization, 4th Ed., Ch. 3.5: "Molecular weight distribution"
❌ "Initiators are catalysts that speed up polymerization."
✅ Initiators are NOT catalysts — they are CONSUMED in the reaction. Each initiator molecule decomposes into 2 radicals, which become the chain ends of polymer molecules. Initiators end up incorporated into the polymer (typically <1 wt%), not regenerated. A catalyst, by definition, is regenerated and not consumed. Confusingly, in some industrial contexts ("PVC catalyst") the word is misused for initiators. True polymerization catalysts (Ziegler-Natta, metallocenes) work differently — they're regenerated and used at very low loading.
📖 Reference: Stevens, M.P. — Polymer Chemistry, 3rd Ed., Ch. 6.2
❌ "Termination always happens by combination — two radicals join into one chain."
✅ Two modes exist: combination (2 radicals join, doubling chain length) and disproportionation (one radical abstracts β-H from other, giving 2 separate chains: one alkane-end, one alkene-end). The dominant mode depends on the monomer: styrene polymers terminate ~95% by combination; MMA radicals (more steric crowding around the radical center) terminate ~75% by disproportionation at 60°C. The ratio shifts with temperature too. This affects DPₙ: pure combination gives DPₙ = 2ν; pure disproportionation gives DPₙ = ν.
📖 Reference: Odian — Principles of Polymerization, 4th Ed., Ch. 3.4d
❌ "Once a chain reaches its 'final' length, it stops growing because it's used up the monomer."
✅ A chain stops growing because of TERMINATION (encounter with another radical) or CHAIN TRANSFER (radical hops to another molecule). NOT because of monomer depletion locally. The growing chain is a tiny radical end on a long inert backbone — it samples bulk monomer through diffusion. At any moment in time, [M] in the bulk is much larger than [P•] (10¹⁰× more), so monomer is essentially infinite. Chain length is set by the COMPETITION between $k_p[M]$ (continue growing) and $k_t[P\bullet]$ (terminate) — pure kinetics, not local depletion.
📖 Reference: Cowie — Polymers: Chemistry & Physics, 3rd Ed., Ch. 6.4
❌ "Chain transfer to monomer is rare and can be ignored."
✅ Even though $C_M$ values are small ($10^{-5}$ to $10^{-3}$), since [M] >> [S] in bulk polymerization, transfer to monomer can be significant. For VAc, $C_M \approx 2 \times 10^{-4}$ — small, but it CAPS the maximum achievable DPₙ at ~5000 even in idealized conditions. Chain transfer to monomer is the universal "background" that limits MW even when no other transfer agent is present. For some monomers (allyl-type), $C_M$ is so large (~$10^{-2}$) that polymerization gives only oligomers.
📖 Reference: Mayo, F.R. — J. Am. Chem. Soc. 65, 2324 (1943); Odian Ch. 3.6
❌ "Branching only happens in step-growth polymers like polyesters."
✅ Free radical polymerization can produce HEAVILY branched polymers via two routes: (1) Intramolecular chain transfer (back-biting) — the active radical loops back to grab an H from its own chain (typical for PE, giving short branches every ~50 carbons); (2) Intermolecular chain transfer — radical abstracts H from a dead chain, then grows as a long branch from that point. LDPE made by radical polymerization at high pressure has 20-30 branches per 1000 C atoms, which is why it's "low density" (entanglement reduces crystallinity). HDPE, made by Ziegler-Natta catalysis, is linear and denser.
📖 Reference: Odian — Principles of Polymerization, 4th Ed., Ch. 3.7: "Chain transfer"
📚 Education Research Sources:
• Coleman, M.M. & Painter, P.C. — Fundamentals of Polymer Science: An Introductory Text, CRC (1998)
• Stadermann, J. & Komber, H. — "Teaching polymer chemistry", J. Chem. Educ. 87, 1175 (2010)
• Pendergrass, B.A. & Olesik, S.V. — "Polymer chemistry in undergraduate curriculum", J. Chem. Educ. 87, 685 (2010)
• Taber, K.S. — Chemical Misconceptions, Vol. II, RSC (2002)