Chapter 21: Reactions of Aromatics

Although benzene, as the prototype of aromatic systems, formally has three C=C double bonds, its reactions are quite different from those of alkenes. Whereas alkenes tend to undergo addition reactions, especially electrophilic additions, benzene tends to under substitution. Essentially the preference for substitution over elimination derives from the importance of retaining the aromaticity of the benzene ring, which would be lost if addition occurred to one of the double bonds.

A Generalized Mechanism for Electrophilic Aromatic Substitution.

q Note that the first step (sometimes called the “attack step”, is rate-determining. That is primarily because it is in this step that the aromaticity of benzene is disrupted. Step 2, in which the stabilizing aromaticity is restored, is naturally fast.

q Also important is the fact the the intermediate carbocation, which in this context is termed an arenium ion, is itself highly delocalized and resonance stabilized, but not aromatic. The tetrahedral carbon to which the electrophile bonded interrupts the efficient cyclic conjugation, since it does not have a 2p_z AO to contribute to the conjugated system. Nevertheless, there are three resonance structures which, somewhat like the benzyl carbocation, distribute the positive charge to the ortho and para positions of the ring.

q We should note also the the carbocation intermediate (the arenium ion) potentially could have reacted with the base/nucleophile, B, at an ortho position, to give a net electrophilic addition reaction, instead of eliminating a proton to yield the substitution product. This kind of Ad_E mechanism would be exactly analogous to the behavior ordinarily observed for alkenes. The arenium ion does not do this because the reaction path involving proton elimination regenerates the aromaticity of the ring, while reaction at the ortho position would have given a product which is not at all aromatic, and would have lost even the resonance stabilization of the arenium ion..

The Reactivity of Benzene.

We have just seen that the qualitative sense of the reactivity of benzene (and other aromatics) is quite different from the reactivity mode preferred in simple alkenes. But what about the quantitative reactivity, i.e., the relative rates of reaction of benzene and, say, ethene as an example of a simple alkene?

q As a direct result of its aromatic resonance stabilization, benzene reacts much more slowly than alkenes with a wide variety of reagents. Recall that stabilization of the reactant tends to increase the activation energy for a reaction and to decrease the rate. A dramatic example is the reaction (or rather the non-reaction) with bromine. Whereas alkenes typically react virtually instantaneously (in real time) with bromine, benzene essentially doesn’t react with bromine at all. [We will see shortly that bromine can be induced to react with benzene by the use of an appropriate electrophilic catalyst, but bromine itself is not reactive enough to undergo reaction with benzene.]

Nitration.

The nitro group (-NO₂) is an extremely versatile substituent. Not only is this substituent important in its own right, but it is readily converted, as we will see later, to a variety of other substituent groups. The ability to introduce this substituent onto the benzene, or other aromatic rings, is useful synthetically as wel as industrially.

q First, recall that the nitro group is a resonance-stabilized group. There are two equivalent resonance structures.

q Also, note that in both structures there is positive charge on nitrogen, but the negative charge is shared equally with the two oxygen atoms. Since both structures are ionic, this group is a very highly polar group. Further, since the aromatic ring will be attached to the positively charge nitrogen atom, the nitro group is a strongly electron-withdrawing group (EWG). In fact, it is the strongest of the substituents commonly encountered inorganic chemistry.

q The nitro group can be introduced , basically, from the nitro portion of nitric acid. However, benzene itself is too unreactive to react directly with nitric acid, requiring a stronger acidic catalyst in order to accelerate the reaction. Typically, sulfuric acid is used as that stronger acid. The nitrating agent HNO₃/H₂SO₄ is often called “mixed acid”.

q The detailed mechanism of the nitration of benzene with mixed acid is given below:

q The usual “two-step mechanism” for aromatic substitution is represented in steps 3 and 4, with the “attack step” being rate determining. The first two steps are necessary in order to generate the “active electrophile”, the nitronium ion. As was pointed out before, nitric acid is not reactive enough to nitrate benzene directly, but must be converted to the stronger electrophile.

Brormination.

Bromination of aromatic rings can also be accomplished efficiently, but, again, molecular bromine is not a reactive enough electrophile to brominate benzene at a convenient rate.

q The most common catalyst for aromatic bromination is the electrophilic catalyst ferric bromide, which is conveniently generated in situ (in place) by the reaction of bromine with catalytic amounts of iron.

q Although, by analogy with nitration, one might conceive that the more reactive electrophile which introduces bromine onto the benzene ring might be Br⁺, this turns out not to be the case. The active brominating agent still contains the catalyst; it is a “complex” of ferric bromide with bromine.

q The detailed mechanism for electrophilic aromatic bromination, catalyzed by ferric bromide is provided below:

q Recall that the halogens, when univalently bonded, have three electron pairs. They are not very basic or nucleophilic, but they are more nucleophilic than basic. Through one of the electron pairs on one of the bromine atoms, a covalent bond is formed to iron through one of the vacant orbitals in Fe⁺³, i.e., with ferric bromide active as the electrophile. This complex is a weak one and is reversibly formed, but the positive charge on bromine makes the terminal bromine of the complex more electron deficient and thus more reactive as an electrophile toward the benzene ring. Incidentally, the central bromine, which also has positive charge, cannot bond to benzene because bromine cannot readily expand its valence to four (this bromine is already trivalent).

q The active electrophile, this complex between the electrophilic catalyst and nucleophilic bromine, is more reactive than bromine but less reactive than free Br⁺ would be. Apparently, rapidly than the hypothetical dissocation of the complex to Br⁺.

q Since these reactions are not carried out in aqueous solution, the best base available to remove the proton from the intermediate arenium ion is FeBr₄^-.

q Incidentally, chlorine reacts with benzene in an entirely similar way. A few very reactive aromatics, which have very electron-rich rings, are able to react directly with molecular bromine or chlorine, without the need for an electrophilic catalyst. A few very unreactive ones may require the generation of Br^++.

Friedel-Crafts Alkylation.

A carbon atom which is attached to a halogen atom(as in an alkyl halide) has electrophilic character, since it represents the positive end of a substantial bond dipole. However, alkyl halides are not very strong electrophiles and cannot, by themselves, effect electrophilic substitution on benzene. However, in a manner very similar to bromination and chlorination, electrophilic catalysts can be used to enhance the fractional positive charge on that carbon atom, and thus to enhance its electrophilicity to the point where it is indeed reactive enough to react with benzene and other aromatics.

q Interestingly, the precise mechanism of Friedel-Crafts alkylation of benzene depends upon the nature of the alkyl group in just the same way as the S_N reactions of alkyl halides with various nucleophiles do.

Mechanism of Alkylation of Benzene with Methyl Halides.

q As in the case of bromination/chlorination, a Lewis acid/Lewis base complex is formed between the electrophilic catalyst, aluminum chloride (recall that Al is in Group III of the Periodic Table and that, like boron, it has a vacant valence shell p orbital when trivalent) and the nucleophilic iodine atom via one of the unshared pairs on iodine.

q This complexation generates positive charge on the iodine atom, which then also attracts electrons from carbon, making the C-I dipole even larger. The now more-electron-deficient methyl group then becomes a stronger electrophile than in the uncomplexed alkyl halide.

q Also as before, the strongest base available to accept the proton from the arenium ion is the negatively charged tetravalent aluminate species.

q IMPORTANT: The above reaction represents an electrophilic substitution (S_E) on benzene, but it is an S_N2 reaction on an alkyl halide. The nucleophile in this latter reaction is benzene, which replaces the iodine substituent of methyl iodide. Step two is a concerted loss of the leaving group AlCl₃I^-, along with the formation of a bond to nucleophlic benzene.

q It is important to understand that the simultaneous existence of electrophilic and nucleophilic substitutions is not by any means an anomaly. The electrophilic and nucleophilic roles are complementary, i.e., the role of an electrophile is to react with a nucleophile, and conversely. So there is always a nucleophilic process that is concomitant with any electrophilic process.

q When benzene undergoes substitution by, say, nitronium ion the organic chemist says this is an electrophilic reaction because from the point of view of the organic molecule it is, i.e., the organic molecule reacts with an electrophile. But, from the point of view of an inorganic chemist, the reaction could be equally well said to be a nucleophilic reaction (of benzene as the nucleophile) with the inorganic nitronium ion.

q The different situation which arises in the Friedel-Crafts alkylation reaction is that both the electrophile and the nucleophile are organic, so that even to the organic chemist the classification as electrophilic or nucleophilic becomes ambiguous. With respect to benzene (i.e., from the point of view of the organic molecule benzene) it is an electrophilic substitution. But, from the point of view of the organic molecule methyl iodide, the reaction is a nucleophilic reaction. In such a case it is necessary to specify the reference molecule: electrophilic with respect to benzene; nucleophilic with respect to methyl iodide.

Friedel-Crafts Alkylation of Benzene with Secondary and Tertiary Halides.

q It should be obvious that tertiary alkyl halides would not undergo substitution by an S_N2 mechanism (steric hindrance in the pentacovalent TS). Instead, they prefer an S_N1 type mechanism involving tertiary carbocations. Since benzene is a very weak nucleophile, secondary halides also react via an S_N1 mechanism.

q The detailed mechanism for the alkylation of benzene by tert-butyl chloride is presented below:

Friedel-Crafts Alkylation By Primary Halides.

The alkylation of benzene by primary halides proceeds in good yields, but is not normally a useful synthetic reaction because it provides mixtures. Although the expected mechanism for a primary halide is the S_N2 mechanism, and this does account for part of the product, a large part (usually the predominant part) of the product results from a rearrangement of the alkyl halide/aluminum chloride complex to give a secondary carbocation, which then proceeds to alkylate benzen by an S_N1 mechanism.

q A typical example is the alkylation of benzene by butyl chloride (1-chlorobutane). The S_N2 mechanism provides butylbenzene (35%), but the rearranged complex affords 1-methylpropylbenzene (sec-butylbenzene).

q Although a primary carbocation is never involved as an intermediate, the Lewis acid/Lewis base complex has sufficient carbocation character on the primary carbon to permit a hydride shift to give the secondary carbocation, which then alkylates benzene via the S_N1 – like mechanism.

Friedel-Crafts Acylation.

Acylation of an aromatic ring, the insertion of an acyl group onto an aromatic ring, is accomplished using an acid halide, normally an acid chloride, in a way analogous to that employed in Friedel-Crafts alkylation.

q Aluminium chloride is again required in order to facilitate the reaction, but in this case the reaction is not catalytic. A stoichiometric amount (equimolar amount) of this promoter is required (we will see why).

q The product is a ketone.

q The mechanism is always of the S_N1 type, because the acylium ion is a highly resonance stabilized cation, having both carbocation and oxonium ion character. The carbocation center is stabilized by one of the electron pairs of the oxygen atom.

q The mechanism of Friedel-Crafts acylation is as follows:

q Although the aluminum chloride is regenerated in step 4 of the mechanism, it immediately forms a rather strong Lewis acid/Lewis base complex with the ketone function. The ketone oxygen is a stronger Lewis base than either the chlorine or the oxygen atom of the acid chloride, so that the aluminium chloride remains complexed to the ketone and is not available for further complexing with the acid chloride. Consequently, a mole of this promoter is needed for every mole of product ketone formed. Incidentally, the ketone is released from the complex upon aqueous workup, because water complexes (and reacts) with aluminum chloride more strongly than does the ketone.

Reaction Rates and Positional Selectivity in the Electrophilic Aromatic Substitution Reactions of Monosubstituted Arenes.

We have just shown how a variety of substituents can be introduced onto a benzene ring using electrophilic aromatic substitution reactions. The products of these reactions are monosubstituted benzene derivatives. Since monosubstituted arenas still have five hydrogens attached to the benzene ring, it is obvious that they may be able to undergo further substitution reactions, successively generating di-, tr-, tetra-, penta-, and even hexasubstituted arenas. We want to consider the second step in this sequence, the conversion of monosubstituted arenas into disubstituted arenas.

q Note that in a monosubstituted arene, the two ortho positions are equivalent as are the two meta positions. There are therefore only three possible disubstitution products (the ipso carbon has no hydrogen attached, and normally reaction would not occur here; in any case this would not give a disubstitution product).

q Besides the positional selectivity problem (ortho vs meta vs para), there is also the question of how the substituent already present on the monosubstituted arene ring affects the overall relative rate of the reaction (compared, e.g., to the rate if that substituent were not present, i.e., compared with the rate of reaction of benzene itself). Some substituents accelerate the rate in comparison to benzene ( in comparison to hydrogen as the “substituent”) and others decelerate the rate. Some substituents are strongly rate accelerating or decelerating and others are weakly so.

q To more rigorously discuss the question of rate effects, we need a general model of the transition state for the rate determining step. Further, to discuss the question of positional selectivity, we need a transition state model for all three transition states, i.e., those leading respectively to ortho, meta, and para disubstituted products. We would then the Method of Competing Transition States to compare the relative rates of these three competing reactions.

q First, we want to develop a generalized TS model for electrophilic aromatic substitution. Recall that the TS can be represented, using resonance theory, as a resonance hybride of reactant-like and product-like structures at the specific geometry of the TS. The dotted line/partial charge structure is then constructed from this and the model characterized as carefully as possible. This includes the use of the Hammond Principle to decide whether the character is extensive (highly developed), moderate, or minimal.

q Since the reaction of the electrophile with the highly stabilized aromatic ring is typically the rate-determining step and is substantially endothermic, the TS resembles the product of that step, which is the arenium ion. Therefore the TS has extensive arenium ion character, which is equivalent to extensive carbocation character at the ortho and para, but not the meta, positions.

q Using this predominant character, we can now reason about the effect that a specific substituent would have on the energy of this TS and therefore upon the rate and selectivity of the relevant reaction of a monosubstituted arene having this substituent present on the ring. Specifically, substituents which stabilize carbocation character will lower the energy of the TS and accelerate the reaction rate. These substituents are called electron-donating groups (EDG’s). Groups which destabilize a carbocation center are electron-withdrawing groups (EWG’s), and these raise the energy of the relevant TS and retard the reaction rate.

q We already know some groups which stabilize carbocation centers (EDG’s). Alkyl groups, in particular, are common but rather effective groups in stabilizing carbocations and carbocation character in a TS. We have also seen that oxygen atoms attached to a carbocation center stabilize that center in a highly effective way, using one of the unshared pairs on the oxygen atom. Nitrogen atoms are potentially even more effective than oxygen at stabilizing carbocation character. Any group which has an unshared electron pair is capable of stabilizing a carbocation center or carbocation character.

Electron-withdrawing Groups.

Electron-withdrawing groups ( EWG’s) are groups which, when attached to a carbocation center, destabilize the carbocation. These are typically inductive effects, because resonance effects always produce stabilization (or no effect), but never destabilization.

Applying the Method of Competing Transition States to the Rationalization of Positional Selectivity in Electrophilic Substitution of Monosubstituted Arenes.

q Consider, first, an electrophilic subsitution reaction (nitration) of toluene as a specific example. Recall that the arenium ion-like TS has extensive carbocation character at the positions ortho and para to the entering electrophile (not to the substituent, the methyl group).

q Use the generalized TS model previously developed, but substitute the nitro group for the general electrophile, E⁺.

q Construct three TS models, one with the entering electrophile para to the substituent methyl group, one with the electrophile meta to the methyl group, and one with the nitro group electrophile ortho to the methyl substituent.

q We see that one TS, the meta TS does not place carbocation character on the ring carbon atom directly bonded to the substituent methyl group. Consequently, the methyl group is unable to stabilize the carbocation center in a major way. The rate of substitution of the nitro group at the meta position is rather similar to that for a position in benzene.

q In contrast, both the para and ortho TS’s generate extensive carbocation character on the ring carbon directly attached to the methyl group, which can then exert a strong stabilizing effect on the TS. The result is that substitution at both the ortho and para positions is accelerated relative to benzene, and also relative to the meta position. Positional selectivity therefore favors o,p over m substitution.

q The results for the nitration of toluene are illustrated below:

q The main result is that ortho and para substitution is favored over meta. This is a general result for any EDG.

q The greater amount of ortho than para product can be explained statistically, since there are two equivalent ortho positions in toluene, but only one para position. On a “per position” basis, the ortho and para sites are equally reactive. Generally, both ortho and para products are formed in major amounts (and would need to be separated), but the meta product is typically rather negligible.

q In terms of reaction rates, toluene reacts about a thousand times faster than benzene. You may have noticed that sulfuric acid (mixed acid) was not used in the results specified above. Toluene is so much more reactive than benzene, that the use of sulfuric acid as a more acidic catalyst is not necessary. Nitric acid is acidic enough to act as the acid catalyst for generating the nitronium ion.

q To compare the relative rate of nitration of toluene with that of benzene, we should use the Method of Competing TS’s. Since meta substitution is only a minor component, most of the rate acceleration comes from reaction at the para and ortho positions. Since the rates of reaction at the latter two positions are about equal, we need only compare one of these with benzene.

q You could go through the same TS arguments for the methoxy or the dimethylamino substituent. This would be good practice. In both of these latter cases, ortho, para substitution is strongly favored and the rate enhancements relative to benzene are even greater. For example, anisole (methoxybenzene) reacts at a rate approximately a million times faster than benzene. In the case of bromination of anisole, no ferric bromide catalyst need be used. Anisole reacts rapidly with molecular bromine, to give the mixture of o- and p-bromoanisole.

The Effect of Electron-Withdrawing Groups (EWG’s) on Positional Selectivity and Relative Rates

q EWG’s exert effects on positional selectivity and relative reaction rates exactly opposite to those for EDG’s: meta substitution is preferred and reaction rates are retarded, usually very strongly retarded.

q The rationalization of the positional selectivity and relative rate effects is parallel to that discussed for EDG’s, except that now the EWG’s destabilize carbocation character and especially strongly so when the carbocation character resides on the ring carbon to which the EWG is directly attached.

q The Method of Competing TS’s can again be used to rationalize positional selectivities. In this case, for brevity, only the para and meta TS’s will be compared, since whatever conclusions apply to the para TS would also apply to the ortho TS.

q We see that the destabilization of the TS is greater for the para (and ortho) TS than for the meta TS, so meta substitution is favored.

q However, the meta TS is also substantially destabilized relative to benzene, because the electrostatic repulsion between the EWG’s dipole and the positive charges on the two nearby, but not directly attached, ring positions is still substantial. Consequently, the rate of reaction, even at the meta position, is much slower than the rate of the corresponding reaction with benzene.

Halogen Substituents.

The halogen substituents essentially are rule-breakers. While the great majority of all substituents conduct themselves according to the rules, i.e., EWG’s are meta-directing and rate-retarding, while EDG’s are o,p-directing and rate enhancing, the halogen substituents are o,p-directing and rate-retarding.

q Since the halogens decrease the rate of electrophilic aromatic substitution, they are by definition EWG’s. In an overall sense they decrease the reaction rate at all positions by destabilizing carbocation character.

q This destabilization occurs as a result of the large C-Hal dipole moment, which has the positive end of the dipole closer to the positive charge on the ring in the TS.

q However, their overall destabilization of the TS by halogens is less at the o,p positions than at the meta position, because they have a modest resonance stabilizing effect on the carbocation character in the arenium ion-like TS when present at the latter positions. Remember that, unlike electrostatic or inductive effects, resonance stabilization of a carbocation center can be effective only when the substituent is directly attached to the center which has carbocation character. Therefore, since positive charge occurs only at the o and p carbons of the ring, only in this position can the halogen exert its resonance stabilizing effect.

q However, the resonance stabilization is less than the inductive destabilization, so that the halogens are overall EWG’s.

q The major products of the nitration of chlorobenzene are the ortho and para isomers.

Conversions of Disubstituted Arenes to Trisubstitued Arenes.

The basic question at issue is as follows. When two substituents are present in the reactant arene, to what position is the entering electrophile preferentially directed? We will consider two characteristic examples.

q Case I.: The two substituents are para to each other, and are of opposite directional tendencies. In the example given, 4-nitrotoluene, the methyl group is an EDG and the nitro group is an EWG. In this case, both groups would prefer for the electrophile to enter at position a , which is meta to the nitro group and ortho to the methyl group. Consequently, there is no problem in predicting the position that the new electrophile will enter. Incidentally, the electrophile could be any of the electrophiles which we have studied in this unit (nitration, bromination, F-C alkylation and acylation, etc.)

q Case II. The two substituents are para to each other and are of the same directional tendency (in this instance, both are o,p-directors). For example, in p-methylphenol, the hydroxyl group would tend to direct the electorphile to position a , while the methyl group would tend to direct it to position b, since both groups are o,p-directors. However, as we have learned, the –OH group is a more powerful EDG than the methyl group, so the electrophile is exclusively directed to position a.