Electrophilic aromatic substitution

Cyclohexene and benzene both have C=C bonds, and in both the π electrons can attack an electrophile. Mix each with bromine, Br₂, at room temperature in the dark. The cyclohexene takes up the bromine at once and becomes 1,2-dibromocyclohexane: the two bromine atoms add across the double bond, as in Addition reactions. The benzene does nothing.

Now add a little iron(III) bromide, FeBr₃, to the benzene. This time the benzene reacts, giving bromobenzene, C₆H₅Br, and HBr. One hydrogen on the ring has been swapped for a bromine, and the ring still has its three double bonds. The bromine has substituted for the hydrogen rather than adding.

This reaction is electrophilic aromatic substitution, or EAS. The name lists its parts: an electrophile reacts with an aromatic ring, and one group replaces another. Five versions of it put five different groups on a ring, and all five follow the same two steps. This page explains why benzene substitutes where an alkene adds, and then how each reaction makes its electrophile.

Two steps: attack, then loss of H⁺

Step 1 is the same move an alkene makes. A pair of π electrons from the ring attacks the electrophile, written E⁺, and forms a new C–E σ bond. The carbon that bonds to E now has four single bonds, so it is sp³. Its neighbor has lost its share of that π bond, so it carries the positive charge.

The cation this makes is the arenium ion. It is also called the sigma complex, because E is joined to the ring by a new σ bond, and the Wheland intermediate, after the chemist who proposed it. All three names mean the same species.

The sp³ carbon has no p orbital, so the ring of overlapping p orbitals is broken. An aromatic ring needs that unbroken ring of p orbitals (Aromaticity), so the arenium ion is not aromatic. Benzene loses its aromatic stabilization, about 36 kcal/mol, in this step.

Step 2 wins that stabilization back. A base removes the H⁺ from the sp³ carbon, and the C–H bonding pair moves into the ring to rebuild the missing π bond. The ring is aromatic again, and E sits where the H used to be.

The base does not need to be strong. It is usually the anion left over from making the electrophile: FeBr₄⁻ after bromination, AlCl₄⁻ after a Friedel–Crafts reaction, and HSO₄⁻ or water after nitration or sulfonation. A weak base is enough for this step.

Why the ring does not add

An alkene's cation ends its reaction by picking up a nucleophile, which bonds to the positive carbon and completes the addition. The arenium ion could do the same. A bromide ion could bond to its positive carbon and give a dibromide.

That dibromide would have two sp³ carbons in its ring. It would stay non-aromatic for good, and the 36 kcal/mol would never come back. Losing H⁺ instead makes the ring aromatic again. The substituted benzene is far lower in energy than the addition product, so substitution wins. Step 1 is the same for an alkene and for benzene; the two part ways at the arenium ion, in step 2.

Where the positive charge sits

Number the ring so that the sp³ carbon is C1. In the first drawing of the arenium ion, the positive charge sits on C2, the carbon next to C1. The C3=C4 π bond beside it can shift toward C2, which moves the charge to C4. The C5=C6 π bond can then shift toward C4, which moves the charge to C6. These three drawings are resonance structures (Resonance), and the real ion is a blend of all three.

Measured from C1, C2 and C6 are the ortho positions and C4 is the para position, the names from IUPAC Nomenclature. The charge never reaches C3 or C5, the two meta positions. Only these three carbons share the positive charge.

Sharing the charge over three carbons makes the arenium ion far more stable than a cation stuck on one carbon, and that is why the reaction can happen at all. It is still much less stable than benzene, because only five of its carbons take part in the π system and the ring is not aromatic. Ortho/meta/para directing effects uses these three positions to predict where a second group goes.

Why step 1 is the slow step

Attacking an electrophile costs benzene its aromaticity, so benzene is a much weaker nucleophile than an alkene. That is why bromine alone left it untouched in the opening example. Every EAS reaction therefore needs a strong electrophile, and most make one in the flask, as the next section shows.

The energy diagram shows the same cost as a tall first barrier. Step 1 climbs to the arenium ion and gives up aromaticity. Step 2 has only a small barrier. The step whose transition state sits highest sets the rate; it is called the rate-determining step (Energy diagrams & the Hammond postulate). For nearly every EAS reaction, it is step 1.

The five reactions, and how each electrophile is made

Every reaction in the table follows the two steps above. What changes is the electrophile, and most of each reagent list is there to make it. FC stands for Friedel–Crafts, after Charles Friedel and James Crafts, who discovered the two reactions in the last two rows.

Halogenation. Take bromination first. Br₂ on its own is not electrophilic enough to react with benzene. FeBr₃ is a Lewis acid (Lewis acids/bases): its iron accepts a lone pair from one bromine of Br₂. That bromine becomes positive and pulls electron density out of the Br–Br bond, which leaves the outer bromine electron-poor enough for the ring to attack.

As the ring bonds to the outer bromine, the Br–Br bond breaks. Its pair stays with the inner bromine, which leaves as part of FeBr₄⁻. That is why the table says the electrophile is Br⁺ "effectively": no free Br⁺ ever forms, but the ring receives a bromine without its electrons.

In step 2, FeBr₄⁻ takes the H⁺, which gives HBr and returns FeBr₃. The same FeBr₃ is used again and again, so a small amount is enough; it is a catalyst. Chlorination works the same way, with Cl₂ and FeCl₃ or AlCl₃.

Fluorine and iodine are the awkward halogens. F₂ reacts too violently to control, and I₂ is too weak an electrophile, so in practice halogenation means chlorination or bromination.

Iodination can still be forced by adding an oxidant, such as HNO₃, H₂O₂ or CuCl₂, along with the I₂. The oxidant converts I₂ into a more electrophilic iodine species, which the ring can attack.

The oxidant has a second job. It removes the HI as fast as the reaction makes it. HI can reduce iodobenzene back to benzene, so without the oxidant the equilibrium lies on the side of the starting materials.

Nitration. Nitration uses a mixture of nitric acid, HNO₃, and sulfuric acid, H₂SO₄. Sulfuric acid is the stronger acid, so it protonates the OH oxygen of nitric acid. That turns the OH into a water molecule still bonded to nitrogen, and water is a good leaving group.

Water then leaves, taking the N–O bonding pair with it, while a lone pair on the negative oxygen moves in to make a second N=O bond. What remains is the nitronium ion, NO₂⁺, written O=N⁺=O. Its nitrogen has two double bonds and no lone pair, so, like CO₂, it is straight, with a 180° angle at N.

The ring attacks the nitrogen. As the new C–N bond forms, one N=O π bond breaks and its pair moves onto oxygen. After step 2, the product is nitrobenzene.

Nitration matters beyond this page. A nitro group can later be reduced to NH₂, which is the usual way to put nitrogen on a ring as an amine (Amines), and an NH₂ on a ring leads on to diazonium chemistry (Aromatic Follow-Through). Nothing on this page depends on that.

Sulfonation. Sulfonation uses fuming sulfuric acid, which is H₂SO₄ with extra sulfur trioxide, SO₃, dissolved in it. SO₃ needs no help to react. Its sulfur is bonded to three oxygens, each pulling electron density away, so the sulfur is strongly δ+. The ring attacks sulfur, and an S=O π bond breaks with its pair moving onto oxygen. After step 2, that oxygen picks up a proton, and the product is benzenesulfonic acid, C₆H₅SO₃H.

Some books write the electrophile as HSO₃⁺, which is SO₃ with a proton on one oxygen. The ring attacks sulfur either way, and the product is the same.

Sulfonation can be run backwards. Heat benzenesulfonic acid in dilute aqueous acid and the SO₃H group comes off, giving benzene back. The reverse reaction is EAS again, with H⁺ as the electrophile. H⁺ adds to the carbon that carries the sulfur, and then SO₃ leaves instead of H⁺.

The amount of water decides the direction. Plenty of water favors the reverse reaction, because water takes up the SO₃ as H₂SO₄; concentrated fuming acid, with little water, favors the forward one.

Some other EAS reactions can reverse too. Friedel–Crafts alkylation, for example, can lose or swap alkyl groups under its own conditions. Sulfonation is the one chemists routinely reverse on purpose, so an SO₃H group can hold a ring position for a while and then be removed. Ortho/meta/para directing effects puts that to use.

Friedel–Crafts alkylation. This reaction puts an alkyl group on the ring, which is why it is called an alkylation. Take 2-chloro-2-methylpropane, (CH₃)₃CCl (tert-butyl chloride), with AlCl₃. The aluminum in AlCl₃ has only six electrons around it, so AlCl₃ is a Lewis acid, and it accepts a lone pair from the chlorine. The C–Cl bond then breaks, and its pair stays on chlorine, which leaves as AlCl₄⁻. What remains is the tert-butyl cation, (CH₃)₃C⁺, a carbon electrophile. This is the same ionization that starts an SN1 reaction (Substitution & Elimination), with AlCl₃ helping the chloride leave.

The ring attacks the cation, and in step 2 AlCl₄⁻ takes the H⁺. That gives tert-butylbenzene and HCl, and returns AlCl₃.

A primary halide such as chloroethane behaves differently. A primary cation is too unstable to form, so the chloroethane–AlCl₃ complex itself is the electrophile. The ring attacks its carbon directly, and AlCl₄⁻ leaves as the new C–C bond forms.

An alkyl halide is not the only way in. An alkene with an acid such as HF or H₃PO₄ gives a carbocation by protonation, and so does an alcohol with H₂SO₄. Either cation alkylates the ring in the same way. Ethylbenzene, the starting material for polystyrene, is made industrially from benzene and ethylene with an acid catalyst, not from chloroethane. Ethylene gives no free primary cation; as with chloroethane, the ring attacks the acid-bound ethylene directly.

Aryl and vinyl halides cannot take part. Their chlorine sits on a carbon of a benzene ring or of a C=C. The cation that would form there has its empty orbital at right angles to the π electrons, so it gets no help from them and is far too unstable to form (Carbocations). Chlorobenzene and vinyl chloride therefore fail as Friedel–Crafts partners, however much AlCl₃ is added.

Friedel–Crafts acylation. An acyl group is a C=O carbon carrying an R group, R–C(=O)–, and an acylation puts one on the ring. Take acetyl chloride, CH₃COCl, with AlCl₃. As before, aluminum accepts a lone pair from the chlorine. Then two pairs move at once. The C–Cl bond breaks and its pair leaves with chlorine as AlCl₄⁻, while a lone pair on oxygen moves in to make a third C–O bond. The result is the acylium ion, CH₃C≡O⁺.

The acylium ion has two resonance structures. In one, CH₃–C⁺=O, the carbon carries the positive charge and has only six electrons. In the other, CH₃–C≡O⁺, an oxygen lone pair has formed a third bond, and every atom has a full octet. That second structure is the major one. Oxygen’s lone pair stabilizes the cation by sharing the positive charge, the same resonance that makes this ion a major fragment of ketones in Mass spectrometry.

Benzene’s π electrons attack the acylium carbon. The C≡O triple bond hands one of its π pairs to oxygen and becomes a C=O. After step 2, the product is acetophenone, a ketone.

Here AlCl₃ is not a catalyst. The ketone’s oxygen is a Lewis base, and it binds AlCl₃ tightly as soon as the product forms. So the reaction needs a full equivalent of AlCl₃ or more, one for each ketone, and water is added at the end to free the ketone from the aluminum.

Three limits of Friedel–Crafts alkylation

Alkylation looks like the direct way to put an alkyl group on a ring. In practice it is the less useful of the two Friedel–Crafts reactions, for three reasons.

The cation can rearrange. Treat benzene with 1-chloropropane and AlCl₃, and the major product is isopropylbenzene, not propylbenzene. The primary propyl cation would carry its charge on C1. A hydrogen on C2 moves over to C1 together with its bonding pair, which is a 1,2-hydride shift (Carbocations). That leaves the positive charge on C2, in the more stable secondary cation, and this is the cation the ring attacks.

Strictly, a free primary cation never forms here. The hydride shifts while the C–Cl bond in the AlCl₃ complex is breaking. Most courses draw a primary cation that then rearranges, and the product is the same either way. Propylbenzene still forms as the minor product, from direct attack on the complex before it rearranges.

The product reacts again. Treat benzene with chloroethane and AlCl₃ and some of the ethylbenzene picks up a second ethyl group, giving diethylbenzenes. An alkyl group pushes electron density into the ring, the same effect that stabilizes a carbocation, so ethylbenzene forms its arenium ion more easily than benzene does. So ethylbenzene reacts faster than benzene, and a ring like this is called activated. The usual fix is a large excess of benzene, so that the electrophile mostly meets benzene rather than product.

Most rings with electron-withdrawing groups do not react. A ring that reacts more slowly than benzene is called deactivated. Nitrobenzene is the standard case: its nitro group pulls electron density out of the ring, which leaves the ring too weak a nucleophile for a Friedel–Crafts electrophile. It is inert enough to serve as a solvent for Friedel–Crafts reactions.

Rings carrying a C=O group such as a ketone, an SO₃H group, a C≡N group, or a positive nitrogen such as NR₃⁺ fail the same way. Halogens are the exception. A halogen also pulls some electron density out of the ring, but it deactivates the ring only weakly, so chlorobenzene and bromobenzene still undergo Friedel–Crafts acylation.

An NH₂ group on the ring fails for a different reason. Its nitrogen lone pair binds AlCl₃, just as the chlorine of the alkyl halide does. That puts a positive nitrogen right on the ring, which pulls electron density out and deactivates it. This third limit, including the NH₂ case, applies to acylation as well.

Acylation avoids the first two

The acylium ion does not rearrange. Its oxygen already stabilizes the positive charge, so no hydride or alkyl shift could give a more stable cation.

Acylation also stops after one substitution. The C=O group of the new ketone pulls electron density out of the ring, so acetophenone is deactivated and reacts more slowly than benzene. A second acylation does not happen.

That makes acylation the way to an unbranched alkyl group. Acylate first, then reduce the ketone’s C=O to CH₂ with a Clemmensen reduction, Zn(Hg) and HCl, or a Wolff–Kishner reduction, hydrazine then hot KOH. For now you need only what they do; Reducing carbonyls, later in the course, shows how. Two steps succeed where one does not.

What carries forward

The arenium ion, with its charge on three carbons, is the starting point of Ortho/meta/para directing effects. A group already on a ring changes both how fast the next substitution goes and where it lands, so the order in which groups go on changes the product.

The reactions on this page are half the toolkit for building substituted benzene rings, which appear in a large share of medicines. The other half is diazonium chemistry, which turns an NH₂ on a ring into many other groups. It is introduced in Amines and completed in Aromatic Follow-Through.