Mechanism of Dehydration of Alcohols

Heating an alcohol with a strong dehydrating acid -- most commonly concentrated

$\text{H}_2\text{SO}_4$, or alternatively passing the alcohol's vapour over hot alumina,

$\text{Al}_2\text{O}_3$ -- removes a molecule of water and forms an alkene. This

acid-catalysed dehydration proceeds by an E1 (unimolecular elimination) mechanism, in three

distinct steps.

Step 1: protonation. A lone pair on the alcohol's oxygen is protonated by the strong acid,

converting the poor leaving group $-\text{OH}$ into the far better leaving group $-\overset{+} {\text{O}}\text{H}_2$ (a neutral water molecule, once it departs, is a much weaker base -- and

hence a much better leaving group -- than hydroxide would be):

Step 2: loss of water to form a carbocation. The protonated alcohol (oxonium ion) then loses

the neutral water molecule to generate a carbocation -- this is the slow, rate-determining step of

the whole mechanism, and, exactly as for the Lucas test, its rate is governed by carbocation

stability: a tertiary alcohol ionises fastest, a primary alcohol slowest. This is why the general

ease of dehydration follows the order $3° > 2° > 1°$, and why primary alcohols typically need

noticeably higher temperatures to dehydrate than tertiary alcohols do.

Step 3: loss of a beta-hydrogen. A base (often another molecule of the alcohol, or

$\text{HSO}_4^-$) removes a hydrogen from a carbon adjacent (beta) to the cationic carbon; the

electron pair from that $\text{C--H}$ bond becomes the new $\pi$ bond, giving the alkene and

regenerating the acid catalyst.

Zaitsev's rule. When more than one beta-hydrogen is available -- as in butan-2-ol, which can

lose a hydrogen from either C-1 or C-3 -- more than one alkene is possible, and the reaction

predominantly gives the more substituted, more stable alkene as the major product (Zaitsev's

rule), because the transition state for its formation is itself more stabilised by the same

hyperconjugative and inductive effects that stabilise the alkene product. Butan-2-ol therefore

gives mainly but-2-ene (more substituted) rather than but-1-ene (less substituted), even though

both are structurally accessible from the same carbocation.

Carbocation rearrangement. Because a genuine carbocation intermediate is formed in step 2, it

can, before it loses a proton, first rearrange to a more stable carbocation by a hydride shift

(a neighbouring $\text{C--H}$'s bonding electrons migrate, with the hydrogen, to the cationic

carbon) or an alkyl shift, whenever such a shift leads to a more stable cation. This is why …