Nitrate toluene and it reacts about 25 times faster than benzene, with the nitro group landing mostly ortho and para to the methyl group. Nitrate nitrobenzene and it reacts far more slowly than benzene, with the second nitro group going meta. In both cases the group already on the ring decided two things: where the new group went, and how fast it got there.
Ortho, meta and para name positions relative to that first group, as in Rings & unsaturation. Ortho is the carbon next to it, meta is one carbon further round, and para is directly across the ring. This page explains which groups send the next electrophile where, and why.
Two questions: how fast, and where
The first question is rate. A substituent that makes the ring react faster than benzene is an activator, or activating group. One that makes it react more slowly is a deactivator.
The second question is position. Chemists call it regiochemistry: which of several possible positions the new group takes. An ortho/para director sends the incoming electrophile mainly to the ortho and para positions, and a meta director sends it mainly to meta.
For most groups the two answers come as a pair. Activators direct ortho/para, and deactivators direct meta. The halogens break the pattern: they slow the ring down, yet they still direct ortho/para. They get their own section below.
Where the positive charge lands
In Electrophilic aromatic substitution, the electrophile adds to one ring carbon and makes the arenium ion. Its positive charge is shared by three ring carbons: the two ortho to the attacked carbon and the one para to it. Now put a substituent, G, on C1, and let the electrophile attack C2, C3 or C4 in turn.
Ortho attack, at C2, puts the charge on C1, C3 and C5. Para attack, at C4, puts it on the same three carbons. Meta attack, at C3, puts it on C2, C4 and C6. So ortho and para attack each give one resonance structure with the positive charge on C1, the carbon that carries G. Meta attack never does.
That one difference decides the direction for every group on this page. If G helps a positive charge on its own carbon, the ortho and para arenium ions gain, and G is an ortho/para director. If G makes that charge worse, the ortho and para ions suffer, and meta is left as the best of the three.
As EAS showed, whatever steadies the arenium ion also lowers the barrier to forming it (the Hammond postulate), so the position with the most stable arenium ion reacts fastest and gives the major product.
Activators: donate, direct ortho/para, speed it up
Take aniline, an NH₂ group on a benzene ring. The nitrogen bonded to the ring carries a lone pair. Follow para attack in the top row of the figure. One resonance structure puts the positive charge on C1, the carbon bonded to nitrogen. The nitrogen then shares its lone pair with C1 and makes a C=N double bond.
That extra structure helps a great deal. In each all-carbon structure, the charged carbon has only six electrons. In the new one, the positive charge sits on nitrogen, and every atom in the ion has a complete octet. Meta attack never puts the charge on C1, so the lone pair has nothing to reach and meta attack gets no such structure. Counting structures would miss this, because every attack has three all-carbon structures. What decides it is the one extra, good structure.
So aniline reacts ortho and para. The same donation also leaves those arenium ions lower in energy than benzene’s own, so aniline reacts faster than benzene. NH₂ is both an ortho/para director and an activator.
The argument works for groups whose first atom, the one bonded to the ring, is nitrogen or oxygen with a lone pair: OH, OR, NH₂, NHR, NR₂, NHCOR, and the phenoxide oxygen, O⁻. Each one donates electron density into the ring through the π system. That is resonance donation. Halogens also carry lone pairs; the section on them below explains why they behave differently.
The strongest of these groups make the ring very reactive. Phenol reacts with bromine water, with no Lewis acid at all, and gives 2,4,6-tribromophenol: every ortho and para position picks up a bromine. Aniline behaves the same way.
Aniline should nitrate fastest of all, yet direct nitration goes badly. Nitration uses concentrated nitric and sulfuric acids. In that acid, the nitrogen’s lone pair picks up a proton, just as ammonia does in Brønsted acids/bases, and aniline becomes the anilinium ion, C₆H₅NH₃⁺.
The –NH₃⁺ group has no lone pair left to donate, and it carries a full positive charge on the atom bonded to the ring. That makes it a deactivating meta director, like the groups in the next section.
The products show both forms reacting. Nitrating aniline directly gives roughly 51% para, 47% meta and only 2% ortho. Much of the para product is usually put down to the small amount of free aniline left in the acid, which reacts far faster than the anilinium ion does. The anilinium ion itself gives mostly meta, though some of the para product comes from it too.
Meta directors: withdraw, direct meta, slow it down
Nitrobenzene shows the opposite case. Nitrate it, and the second nitro group goes mostly meta, giving 1,3-dinitrobenzene. In the nitro group, the nitrogen carries a formal positive charge, and that nitrogen is bonded directly to the ring.
Now follow ortho attack in the figure. One of its resonance structures puts the ring’s positive charge on C1, right next to that positive nitrogen. Two positive charges on neighboring atoms repel each other strongly, so this structure is very high in energy, and the whole ortho arenium ion suffers for it. Para attack makes the same bad structure. Meta attack puts the charge on C2, C4 and C6, next to C1 but never on it.
So the nitro group does not stabilize meta attack. Meta is simply the least bad of three bad choices. The nitro group also pulls electron density out of the whole ring, by induction and by resonance, so every position reacts slowly. Nitrobenzene nitrates only 10⁻⁷ to 10⁻⁸ times as fast as benzene. Every position reacts slowly, and the little reaction that does occur goes meta.
The same reasoning covers every group whose first atom carries a full or partial positive charge: NO₂, CHO, COR, COOR, COOH, CN, SO₃H, CF₃ and NR₃⁺. In a carbonyl or nitrile group, the carbon bonded to the ring is δ+ because oxygen or nitrogen pulls electrons away from it. In CF₃, three fluorines do the same to their carbon. In SO₃H, the sulfur bonded to the ring carries three electron-pulling oxygens. All of these are deactivating meta directors.
Halogens: they slow the ring but still direct ortho/para
Chlorobenzene nitrates about thirty times more slowly than benzene, yet the new nitro group still goes ortho and para. So chlorine is a deactivator and an ortho/para director at the same time. Two effects are at work, and each one decides a different question.
First, chlorine is electronegative, so it pulls electron density out of the ring through the σ bond. This is induction, the same pull that makes chloroacetic acid stronger than acetic acid in Factors affecting acidity. It leaves every ring position poorer in electrons, so the whole ring reacts more slowly.
Second, chlorine has lone pairs, and it can donate one into the ring by resonance, just as NH₂ does. That help reaches only the ortho and para arenium ions, exactly as in the aniline figure, with a C=Cl⁺ bond in place of C=N⁺.
Induction is the larger effect overall, so it sets the rate, and the ring is deactivated. But induction slows all three positions, while resonance still helps only ortho and para. So resonance sets the position.
Why is chlorine’s donation so much weaker than nitrogen’s? Its lone pairs sit in 3p orbitals, which are larger than carbon’s 2p and overlap with it poorly. The same mismatch makes the chlorine of an acid chloride a weak donor to its C=O, which is why an acid chloride’s C=O band sits highest in IR.
Fluorine is the exception within its own family. Its lone pairs sit in 2p orbitals, the same size as carbon’s, so it is by far the best resonance donor of the four halogens. As a result, fluorobenzene is among the least deactivated halobenzenes: it reacts roughly as fast as iodobenzene, several times more slowly than benzene. Chlorobenzene and bromobenzene are slower again, by a further factor of about five.
Fluorobenzene is also the most strongly para-selective of the four. Induction fades quickly with distance, so fluorine’s strong pull falls hardest on the nearby ortho carbons. By the para carbon the pull has mostly faded, while the resonance donation still arrives in full.
| Substituent | Effect on rate | Directs |
|---|---|---|
| –NH₂, –NR₂, –OH, –O⁻ | strongly activating | ortho/para |
| –OR, –NHCOR | activating | ortho/para |
| –R, –Ar | weakly activating | ortho/para |
| –F, –Cl, –Br, –I | weakly deactivating | ortho/para |
| –CHO, –COR, –COOR, –COOH, –CN, –SO₃H | deactivating | meta |
| –NO₂, –NR₃⁺, –CF₃ | strongly deactivating | meta |
Ortho versus para
An ortho/para director gives both products, so which one wins? Nitration of toluene gives about 58% ortho, 37% para and 4% meta. There are two ortho positions and only one para, so chance alone would predict two ortho for every para. A methyl group is small, so the electrophile can attack right beside it, and the ratio comes close to that.
Swap the methyl for a tert-butyl group, a carbon bearing three methyl groups, and the ratio turns around. Nitration of tert-butylbenzene gives about 75% para and only 16% ortho. An electrophile attacking an ortho carbon now has to squeeze in beside that bulky group.
So the ortho-to-para ratio mostly reflects the size of the group already on the ring. The ortho and para products can be separated, and many industrial processes are designed to favor one of them.
Two groups on one ring
In 4-methylanisole, a methoxy group (OCH₃) sits on C1 and a methyl group on C4. Each group’s para position is taken by the other, so only the ortho positions are left. OCH₃ directs to C2 and C6, while CH₃ directs to C3 and C5. The two groups disagree, and OCH₃ wins because it is the far stronger activator: bromination puts the bromine on C2, next to the methoxy group.
That is the general rule. When two groups disagree, the stronger activator controls the position. An activator beats a deactivator, and a strong activator beats a weak one. The table above gives the order.
When the groups agree, crowding has the last word. In m-xylene (1,3-dimethylbenzene), both methyls direct to C2, C4 and C6. C2 is ortho to both methyls, so on electronic grounds alone it looks like the best site. In practice, nitration puts most of the nitro group at C4 (C6 is the same position by symmetry), and only a small share at C2, because an electrophile struggles to fit into a position wedged between two groups. So check the electronic effects first, and then check whether the winning position is wedged.
Planning the order of steps
Target: m-bromonitrobenzene, a bromine and a nitro group meta to each other, made from benzene.
Nitrate first, then brominate. The nitro group is a meta director, so the bromine goes meta. This gives the target. ✓
Brominate first, then nitrate. Bromine is an ortho/para director, so the nitro group goes ortho or para. This gives o- and p-bromonitrobenzene, not the meta isomer. ✗
The two routes use the same two reactions and the same reagents, and only one order gives the target. To make the ortho or para isomer instead, you would deliberately brominate first.
Two practical tricks
Moderate an amine by acylation. Aniline is too reactive for controlled substitution, and its nitrogen binds the Lewis acid that a Friedel–Crafts reaction needs. Acetic anhydride turns aniline into an amide, acetanilide (Acid chlorides & anhydrides, later in the course, shows how this reaction works). The nitrogen now shares its lone pair with the carbonyl, so it is much less basic and a much milder donor, but it still directs ortho/para. Hydrolysis removes the acetyl group afterwards. The amide is put on, the ring reacts, and the amide comes off again.
Block a position with sulfonation. Sulfonation is reversible, as Electrophilic aromatic substitution showed. So a sulfonic acid group (SO₃H) can be put on the para position just to occupy it. The reaction you want then runs at ortho, and hot dilute aqueous acid takes the SO₃H group off again. A group used this way, only to control where another group goes, is a blocking group.
Target: 2-nitroaniline, an NH₂ and an NO₂ ortho to each other, made from aniline.
Why not nitrate aniline directly? The anilinium ion and the oxidation described above would give a mixture rich in the wrong isomer, in poor yield.
Step 1. Moderate the nitrogen. Acetic anhydride converts aniline into acetanilide. As an amide, it stays an ortho/para director in the strong acid of the next two steps.
Step 2. Block the para position. Left alone, acetanilide would nitrate mostly para, because the bulky amide group crowds its ortho positions. So fuming sulfuric acid (H₂SO₄ with dissolved SO₃) first puts SO₃H on the para carbon, C4.
Step 3. Nitrate. With C4 occupied, HNO₃/H₂SO₄ puts the nitro group on C2. Both groups agree here: C2 is ortho to the amide nitrogen and meta to SO₃H, a meta director.
Step 4. Take both temporary groups off. Heating in dilute aqueous sulfuric acid reverses the sulfonation and hydrolyzes the amide back to the amine (Nucleophilic acyl substitution, later in the course). Neutralizing the acid then lets the 2-nitroaniline be collected.
Only one of the four steps, the nitration, adds a group that stays in the product. The other three steps only put temporary groups on or take them off. Aromatic syntheses often spend most of their steps this way.
Preview: aniline itself is made from benzene by nitration followed by reduction of the nitro group to NH₂. Amines covers that reduction.
What carries forward
Directing effects are what make substituted aromatic synthesis possible. Diazonium chemistry, introduced in Amines and completed in Aromatic Follow-Through, will add ways to reach substitution patterns that directing effects alone cannot. The reasoning here is the one used for carbocations and conjugated systems throughout the course: draw the intermediates, find the most stable one, and the reaction goes that way.