When the Elimination Has a Choice
By now you know the mechanics of elimination cold: a base plucks a beta-hydrogen — a hydrogen on the carbon NEXT to the leaving group — while the leaving group departs, and the two carbons in between fold together into a carbon-carbon double bond. In the E2 this happens in one anti-periplanar shove; in the E1 it happens in two steps through a carbocation. But here is the wrinkle the earlier guides set aside: many substrates have more than one DIFFERENT beta-carbon, each carrying its own beta-hydrogens. Pluck from one neighbour and you get one alkene; pluck from the other and you get a different alkene. Same molecule, same leaving group, two possible products.
When a reaction can give more than one structural product and PREFERS one over the others, chemists call that preference regioselectivity — literally selectivity about which REGION of the molecule reacts. (You met its sibling already: Markovnikov's rule was the regioselectivity of addition, picking which carbon gets the proton.) Elimination has its own version of the question, just pointed the other way: not where the proton lands, but which beta-hydrogen leaves. The rest of this guide is the answer — and the honest news that the answer flips depending on the base you choose.
CH3 base plucks a beta-H...
| ...from the LEFT neighbour or the RIGHT?
CH3 - CH2 - C - CH3
|
Br <- leaving group
pluck LEFT (CH2) -> CH3-CH=C(CH3)-CH3 trisubstituted (Zaitsev, more stable)
pluck RIGHT (CH3) -> CH2=C(CH3)-CH2-CH3 disubstituted (Hofmann, less stable)Zaitsev: The Stable, Crowded Winner
The default answer, true most of the time, is Zaitsev's rule: elimination favours the MORE substituted alkene — the one with more carbon groups hanging off the two double-bond carbons. In the sketch above, plucking the left CH2 gives a trisubstituted alkene (three carbons attached to the C=C), and that is the major product. Why? Because Zaitsev's rule is really a statement about alkene stability, not about hydrogens at all. The more alkyl groups you bolt onto a double bond, the lower its energy — partly because those neighbouring C-H bonds lean in and share electron density with the pi bond, the same hyperconjugation that stabilised carbocations in the mechanisms rung, and partly because the bulky groups sit more comfortably spread around a flat double bond than crammed onto a single bond.
So the stability ladder for simple alkenes runs: tetrasubstituted > trisubstituted > disubstituted > monosubstituted, with the trans (E) form of any disubstituted alkene a notch more stable than its cis (Z) twin because the two big groups sit on opposite sides instead of bumping. When two product alkenes are racing, the lower-energy one usually has the lower-energy transition state leading to it (a Hammond-style argument from the energy-diagrams guide), so it forms faster AND in greater amount. That is the whole engine behind 'more substituted wins.'
Hofmann: When the Lean Alkene Wins
Now the twist. Swap your small base for a big, fat one — most famously potassium tert-butoxide, ((CH3)3CO-K+), whose three methyl arms make it a clumsy giant — and the preference can flip to the LESS substituted alkene. This reversed outcome is the Hofmann rule. The cause is not some new electronic principle; it is pure geometry, the same steric bulk story you saw strangle the SN2. The Zaitsev product comes from plucking a hydrogen on a CROWDED, more-substituted beta-carbon — the very carbon already ringed with alkyl groups. A bulky base cannot wedge its way in there. It can, however, easily reach the lonely hydrogens on the LEAST crowded beta-carbon (a terminal CH3, all hydrogens and elbow room), and so it preferentially removes those, giving the less substituted alkene.
There is a second, historically older road to the Hofmann product: a bulky, charged LEAVING GROUP instead of a bulky base. This is the classic Hofmann elimination, where an amine is first methylated to a quaternary ammonium salt (-N+(CH3)3), turning a hopeless leaving group into a workable one. That trimethylammonium group is enormous and positively charged; it crowds the more substituted side and electronically biases the system, so even an ordinary base gives mostly the less substituted alkene. Same destination — the lean alkene — reached either by making the BASE too big or by making the LEAVING GROUP too big. Either way, the bottleneck is steric access to the beta-hydrogen.
Steering the Product on Purpose
This is the part worth carrying away: the base is a STEERING WHEEL, not just an on-switch. If you want the Zaitsev product, reach for a small, compact base that can slip in anywhere — hydroxide (HO-), ethoxide (CH3CH2O-), or methoxide — and let stability call the shot. If you want the Hofmann product, reach for a fat, hindered base — tert-butoxide is the workhorse — and let geometry override stability. Same substrate, same E2 mechanism, opposite major alkene, chosen deliberately by which base you drop into the flask.
- Draw the substrate and circle the leaving group. Then find every DIFFERENT beta-carbon (each carbon directly bonded to the leaving-group carbon that still bears at least one hydrogen).
- For each beta-carbon, imagine removing one of its hydrogens and forming the C=C. Sketch the alkene you would get and count how many carbon groups sit on the two double-bond carbons.
- Rank those candidate alkenes by stability: more substituted is more stable, and trans beats cis. The most substituted one is the Zaitsev product; the least substituted is the Hofmann product.
- Look at the base. Small base (or E1 conditions) => Zaitsev usually dominates. Bulky base (or a bulky charged leaving group like -N+(CH3)3) => Hofmann tilts to the front.
One honest caveat about E1, which runs through a flat carbocation rather than a one-step E2. Because the proton comes off in a SEPARATE, later step from a fully formed cation, the deciding factor is purely the stability of the alkene being born — there is no bulky base squeezing in at the key moment. So E1 almost always gives the Zaitsev product, regardless of base size. The Zaitsev-versus-Hofmann steering wheel is really an E2 phenomenon; in E1, stability wins by default.
Honest Limits and a Bridge Ahead
A few caveats keep this honest rather than slick. First, these rules give you the MAJOR product, not the only one — real flasks return mixtures, often 70:30 or 80:20 rather than 100:0, and you predict where the balance leans, not a single clean answer. Second, the E2 has a geometric requirement you met earlier: the beta-hydrogen and the leaving group must be anti-periplanar, lined up on opposite sides of the bond. In rigid rings, that requirement can FORCE a non-Zaitsev product simply because the 'right' hydrogen physically cannot get into the anti position — a reminder that conformation can outrank both Zaitsev and Hofmann. Third, temperature nudges things too: more heat generally favours elimination over substitution overall, a knob you will turn in the final guide of this rung.
Step back and the pattern is satisfying. Regioselectivity in elimination is not a pair of arbitrary names to memorise; it is a single tug-of-war between two forces you already understand — thermodynamic stability pulling toward the crowded alkene, and steric access pulling toward the open one. A small base lets stability win (Zaitsev); a bulky base or bulky leaving group lets access win (Hofmann). That same stability-versus-sterics tension ran through SN2 versus SN1, and it will run through the four-way substitution-elimination fight you settle next. Learn to feel which force is bigger in a given flask, and you are no longer memorising rules — you are reading the molecule.