A different way to break a bond
Look back over everything you have built so far in this ladder and one habit runs through all of it: bonds break *unevenly*. In an SN2, an E2, an electrophilic addition, the two electrons of a bond always leave together, both going to one atom — so you get a carbocation and a leaving group, or a base and a proton, one positive partner and one negative. That uneven split is heterolysis, and it is the whole reason the curved-arrow language of nucleophiles and electrophiles works: an electron pair flows from a rich site to a poor one. Welcome to the chapter where that habit breaks.
The other way to break a bond is to split it fairly. In homolysis, the two shared electrons part like a divorce settlement — one electron to each fragment. Neither piece ends up charged. Instead each carries a single, *unpaired* electron, and that lonely electron is the whole story: a fragment with an unpaired electron is a free radical. It is not positive like a carbocation, not negative like a carbanion — it is electrically neutral but electronically restless, with an odd electron count and a powerful urge to find a partner for that lone electron. Most radicals live for microseconds. They are intermediates, not products you bottle.
The fishhook: one electron, one barb
All this chapter you have drawn full curved arrows, each one moving a *pair* of electrons from tail to head. Radical mechanisms need a new pen. A fishhook arrow has only a single barb, and it moves exactly *one* electron. Forming a bond between two radicals therefore takes two fishhooks, one from each lone electron, curving together to make the new pair. Homolysing a bond takes two fishhooks again, this time pointing apart — one electron peeling off to each fragment. The rule of thumb is dead simple: full arrow means a pair, fishhook means a single. Mixing them up is the most common beginner's slip in this whole topic.
ELECTRON-PAIR (ionic) world: SINGLE-ELECTRON (radical) world:
full curved arrow = 2 e- fishhook arrow = 1 e-
heterolysis: homolysis:
A:B -> A+ + :B- A:B -> A. + B.
(pair goes to B) (one electron to each)
intermediates: cation / anion intermediate: neutral radical (odd e-)Here is the deeper reason this matters, beyond bookkeeping. Because a radical carries one electron rather than a charge, it does not care much about solvent polarity, it is not pulled toward acids or bases, and it is not stabilised by the things that stabilise ions. A carbocation loves a polar solvent that can wrap around its charge; a radical shrugs at solvent entirely. So radical reactions often run cleanly in non-polar solvents, even in the gas phase, exactly where ionic reactions stall. Different intermediate, different rules — that is why this chapter feels like a separate country.
How easily a bond breaks: BDE and radical stability
Which bond homolyses, and which radical forms? The single most useful number is the bond dissociation energy (BDE): the energy needed to split one specific bond evenly into two radicals. Weak bonds break first. A Cl–Cl bond (about 240 kJ/mol) gives way to a flash of UV light; a strong C–H bond (often above 400 kJ/mol) survives the same flash untouched. This is exactly why, in halogenation, the *halogen* breaks while the alkane sits still — the energy you pour in simply finds the cheapest bond to homolyse. Read a table of BDEs and you can predict, with no hand-waving, where any radical reaction will start.
BDEs also reveal which carbon radicals are *stable*. Compare the energy to pull a hydrogen off the end of propane versus off its middle: the C–H bond at the middle carbon is weaker, because the radical it leaves behind is more stable, so less energy is stored in breaking it. Run this comparison across many molecules and a clear ladder appears — a radical is more stable the more carbon groups surround its lonely electron, the central fact this whole rung leans on. The order is tertiary > secondary > primary > methyl, the very same ranking you learned for carbocations. That echo is not a coincidence.
Why the same order for a positive ion and a neutral radical? Both have an *electron-deficient* carbon — a cation is short a full pair, a radical is short just one — and both are helped by the same trick, hyperconjugation: neighbouring C–H and C–C bonds donate a little electron density into the empty or half-empty orbital, smearing out the deficiency. The order runs 3deg > 2deg > 1deg > methyl: more alkyl neighbours, more smearing, more stability. A radical attached to a pi system does even better, because the lone electron can delocalise over the whole conjugated framework — which is exactly why allylic and benzylic positions, met later in this rung, are so reactive.
The chain: initiation, propagation, termination
Radicals rarely act once and stop. Because each radical is desperate to pair up, when it reacts it usually grabs an electron from a closed-shell molecule — and in doing so it leaves a *new* radical behind. That handoff is the engine of a chain mechanism, and every radical chain you will ever meet is built from three kinds of step. Get these three roles straight and you can read any radical mechanism on the page.
- Initiation creates the first radicals from neutral molecules, usually by homolysis driven by light or heat, or by a special starter molecule with a weak bond (a peroxide, R–O–O–R, snaps apart easily). It is expensive and rare — it only needs to happen once to light the fuse, and it makes no product on its own.
- Propagation is where the work happens. A radical reacts with a closed-shell molecule and hands the radical character on to a fresh species — one radical in, one radical out. These steps repeat in a loop, each lap making a little product, so they must conserve the radical: count one unpaired electron on the left and one on the right of every propagation step.
- Termination destroys radicals — two of them collide and pair their lone electrons into a new bond, leaving only closed-shell molecules. Each termination removes two radicals and snaps that strand of the chain. Because radicals are scarce, two of them meeting is rare, so a single initiation can drive thousands of propagation laps before a chance termination ends the run.
The magic lives entirely in propagation. Each lap consumes one radical and produces another, so the radical is never used up — it is passed around the loop like a relay baton, making a sliver of product on every pass. That is why one photon of initiation can turn over many thousands of molecules: the chain feeds itself. It is also why a trace of impurity that mops up radicals (an *inhibitor*) can shut a reaction down completely, far out of proportion to how little of it you added — kill a few baton-carriers and every strand they would have run goes dark. This self-feeding turnover is the defining feature of any chain mechanism.
Why chains matter — and where they bite
This is not a museum reaction. The single most important radical chain on Earth is radical polymerization: an initiator radical adds to the double bond of a small molecule like ethylene, CH2=CH2, making a new, longer radical that adds to the next ethylene, and the next, stitching thousands of units into one giant chain. Each addition is a propagation step that hands the radical down the growing polymer; termination caps it off. The polyethylene in a milk jug, the polystyrene in a coffee lid, the acrylics in paint — all are chains of small alkenes zipped together by a radical relay. The same self-feeding loop you just traced makes most of the plastic you touch.
Be honest about the limits, because they are real and they shape everything ahead. Radicals are hard to *control*: with little charge to steer them and no leaving group to invert, they are blunter than the precise ionic reactions you mastered earlier. A free radical does not give you clean stereochemistry the way an SN2 inverts a centre — the radical's geometry is nearly flat and easily scrambled, so a chiral centre usually races. Selectivity has to be coaxed (the bromine-vs-chlorine pickiness you saw, or a special reagent like NBS for allylic bromination). The next guides in this rung answer two questions this one raises: how do we steer radicals well, and how do *metals* let us make bonds with surgical precision that neither ions nor radicals can match?