Ethylamine, CH3CH2NH2, reacts with bromoethane, with acetyl chloride and with acetone. Each time, the lone pair on its nitrogen attacks a carbon. That lone pair makes an amine a good nucleophile as well as a good base (Structure & basicity).
The three reactions differ in what happens after that first attack. After an alkylation, the product can attack again, and again. After an acylation, it cannot. With acetone, the result depends on what else is in the flask. This page is about that difference: how to make one new bond to nitrogen and then stop.
SN2 alkylation: a messy way to build amines
Mix ethylamine with bromoethane. The nitrogen lone pair attacks the CH2 carbon of bromoethane and bromide leaves, in an SN2 reaction (Substitution & Elimination). The nitrogen now carries two ethyl groups and a positive charge. A second molecule of ethylamine takes one of its N–H protons, and diethylamine is left.
That looks like a clean route to diethylamine, but diethylamine is a nucleophile too. Each ethyl group pushes electron density onto the nitrogen, so diethylamine attacks bromoethane faster than ethylamine does, and it competes for the bromoethane still in the flask. It picks up a third ethyl group, which gives triethylamine. Triethylamine is more crowded and reacts more slowly, but it still reacts, and it gives the tetraethylammonium ion, (CH3CH2)4N+. This quaternary ammonium salt has four carbons on its nitrogen and no lone pair, so the chain stops there.
Direct alkylation of ammonia or a primary amine therefore gives a mixture of the mono-, di-, tri- and tetra-alkylated products. The mixture is hard to separate, and much of the starting material is wasted. Adding only one equivalent of bromoethane does not help, because the products compete with the starting amine for it.
There are two ways to deal with it. The first is to let the runaway finish. Exhaustive alkylation means adding excess alkyl halide deliberately, so that every molecule goes all the way to the quaternary salt. That is fine whenever the salt is the product you want.
A quaternary salt made this way is also the starting point of Hofmann elimination, later in this chapter, which removes the nitrogen and leaves an alkene.
The second way is to alkylate a nitrogen that can react only once. The Gabriel synthesis does this with phthalimide, which Making amines covers in full. Phthalimide is a benzene ring fused to a five-membered ring, and its nitrogen sits between two C=O groups. Those two C=O groups make its N–H acidic enough for KOH to remove. The anion does one SN2. Afterwards the nitrogen has no H left, and its lone pair is shared with both C=O groups, so it cannot be alkylated a second time.
Hydrolysis then releases a clean primary amine. The Gabriel synthesis works by making the first product unable to react again, and that idea runs through the rest of this page.
Acylation: the clean, self-terminating alternative
Ethylamine and acetyl chloride give N-ethylacetamide, CH3CONHCH2CH3, an amide. The nitrogen lone pair adds to the C=O carbon, the C=O re-forms, and chloride leaves. This addition–elimination is the mechanism taught in Nucleophilic acyl substitution. An anhydride such as acetic anhydride gives the same amide, with acetic acid as the by-product instead of HCl.
The key change is at the nitrogen. In the amide, its lone pair is shared with the new carbonyl group. This is the same resonance that makes an amide such a weak base (Structure & basicity), and it leaves the amide nitrogen almost no nucleophilicity and almost no basicity.
The reaction therefore self-terminates. After one acylation the nitrogen is switched off, so there is no mixture and no over-reaction.
Acylation with an acid chloride also releases HCl. Left alone, that acid would protonate the remaining ethylamine, and an ammonium ion has no lone pair, so the reaction would stop at 50% conversion. A base is added to take up the HCl. It can be a second equivalent of the amine itself, or a tertiary amine such as triethylamine, Et3N. A tertiary amine has no N–H, so it cannot end up as an amide; it just takes the proton. When an even less nucleophilic base is needed, chemists use Hünig’s base, iPr2NEt (iPr is an isopropyl group, (CH3)2CH), whose two isopropyl groups crowd the nitrogen.
Pyridine takes up the HCl as well, and it also speeds the acylation up. Acid chlorides & anhydrides showed how: pyridine attacks the acid chloride first, and the N-acylpyridinium ion it forms, which is more reactive than the acid chloride, passes the acyl group on to the amine. So pyridine is not a non-nucleophilic base: here it is the nucleophile that carries the acyl group. DMAP, 4-dimethylaminopyridine, runs the same relay even faster.
Reductive amination: the reliable route
Acetone and methylamine, with sodium cyanoborohydride (NaBH3CN) in the flask, give N-methylpropan-2-amine, (CH3)2CH–NH–CH3. The nitrogen ends up on the old carbonyl carbon. This is reductive amination: an amine and an aldehyde or ketone condense to a C=N compound, and that C=N is reduced to a C–N single bond in the same flask.
The condensation is the one in Imines and enamines. The amine adds to the C=O, the carbinolamine it forms (rarely isolated) loses water with help from acid, and an iminium ion results: a C=N with a positive charge on nitrogen. Every step is reversible. Imine formation runs fastest in mildly acidic solution, as that page explains, so reductive amination is run under the same conditions.
The reduction is a hydride addition. A hydride from boron bonds to the iminium carbon, and the C=N π electrons move onto the nitrogen, which gives the amine.
Primary and secondary amines both work. A primary amine’s iminium ion can lose its N–H proton to give the neutral imine, but in the acidic solution the imine picks up a proton again, and the iminium ion is what gets reduced. A secondary amine’s iminium ion has no N–H to lose. On its own it would go on to an enamine, but here it is reduced as well, and the product is a tertiary amine.
The reducing agent matters. NaBH4 (Reducing carbonyls) would reduce the ketone to an alcohol before it had a chance to condense. In NaBH3CN, the electron-withdrawing cyano group on boron weakens the hydride. NaBH3CN reduces the far more electrophilic iminium ion much faster than it reduces the ketone, so the condensation and the reduction can run in the same flask at the same time. Sodium triacetoxyborohydride, NaBH(OAc)3, works the same way and is now the more common choice in practice. Hydrogen gas over a metal catalyst, such as Pd or Ni, can also reduce the C=N. Reagents like these two are called attenuated hydride sources, meaning weakened on purpose.
Reductive amination is easier to control than direct alkylation, for three reasons. The first condensation is fast: the primary amine and the carbonyl form the C=N quickly. The product, a secondary amine, is more crowded and can form only a charged iminium ion, so it condenses with a second carbonyl more slowly. And the reduction is not undone: once the C=N has become a C–N single bond, the product is committed. Dialkylation is still a known side reaction, worst with ammonia or with unhindered aldehydes such as formaldehyde. Forming the imine first and adding the reducing agent afterwards keeps it small.
The target is a cyclohexane ring carrying an NH group, with an ethyl group on the nitrogen.
Route A, alkylation: cyclohexylamine and bromoethane. This gives a mixture of the secondary amine, the tertiary amine, the quaternary salt and unreacted starting material. The yield is poor and the separation is hard.
Route B, reductive amination: cyclohexylamine and acetaldehyde, CH3CHO, with NaBH3CN. The iminium ion forms and is reduced, and the main product is N-ethylcyclohexylamine. Acetaldehyde is small and unhindered, so some of the product can condense with a second molecule and go on to the tertiary amine. This is the case where forming the imine first pays off.
The same bond forms from the same fragments, and only one route is practical. This is why reductive amination is one of the most-used reactions in medicinal chemistry.
Give the product when benzylamine, PhCH2NH2, is treated with each of: (a) acetyl chloride and Et3N; (b) excess CH3I with K2CO3; (c) acetone and NaBH3CN; (d) acetic acid. Ph is a phenyl group, a benzene ring bonded through one of its carbons.
(a) Acetyl chloride + Et3N. Nucleophilic acyl substitution. The nitrogen attacks the carbonyl carbon, chloride leaves, and Et3N takes the proton from nitrogen. The product is PhCH2NH–C(=O)CH3, N-benzylacetamide. It stops there, because the amide nitrogen has no nucleophilic lone pair left.
(b) Excess CH3I, with K2CO3 to take up the HI. Here the chain runs to the end. Three SN2 alkylations in a row give the quaternary salt PhCH2N(CH3)3+ I−. With a limited amount of CH3I you would get a mixture instead.
(c) Acetone + NaBH3CN. Reductive amination. The amine condenses with acetone to the C=N compound PhCH2N=C(CH3)2, and that C=N is reduced. The product is PhCH2NH–CH(CH3)2, a secondary amine. Notice that the new C–N bond is on the old carbonyl carbon.
(d) Acetic acid. A trap: no amide forms. The amine is a base and acetic acid is an acid, so all that happens is a proton transfer, giving the salt PhCH2NH3+ −O2CCH3. To reach the amide you would have to heat that salt above about 150 °C or, in practice, activate the acid first.
Nitrous acid and diazonium salts
Aniline, the simplest aromatic amine, reacts with sodium nitrite (NaNO2) and hydrochloric acid at 0 °C to give benzenediazonium chloride, C6H5–N2+ Cl−. Its NH2 group has become an N2+ group. A salt of this kind, Ar–N2+ X−, is an aryl diazonium salt. Ar stands for an aryl group: an aromatic ring, such as a benzene ring, bonded through one of its ring carbons, and X− is the negative ion that balances the charge.
The reaction is ordinary amine nucleophilicity at work. The acid turns NaNO2 into nitrous acid, HO–N=O, and then protonates it, so that it loses water and gives the nitrosonium ion, N≡O+, a small, linear electrophile. The amine’s lone pair attacks the nitrogen of NO+, and the amine nitrogen loses a proton, which gives an N-nitrosamine, Ar–NH–N=O. In a primary amine, the hydrogen still left on that nitrogen then moves to the oxygen, by two proton transfers, giving Ar–N=N–OH. The acid protonates that OH, and it leaves as water. What remains is the N2+ group. Diazonium salts & Sandmeyer draws every curved arrow of this mechanism.
The flask is kept at 0–5 °C because the diazonium salt decomposes above about 5 °C, giving off nitrogen gas, and the product is lost with it.
What nitrous acid does depends on the class of amine. There are three outcomes.
- A primary aromatic amine gives the aryl diazonium salt described above. It is the useful case.
- A primary alkyl amine gives an alkyl diazonium ion, R–N2+, which loses N2 almost as soon as it forms. The carbocation left behind (Carbocations) gives a mixture of substitution, elimination and rearrangement products, so the reaction is rarely useful for synthesis. Its classic use is analytical: measuring the volume of nitrogen gas given off.
- A secondary amine, alkyl or aromatic, stops at the N-nitrosamine, R2N–N=O. After nitrosation its nitrogen has no hydrogen left, so it cannot go on to N2+. N-Nitrosamines are yellow compounds and potent carcinogens, which is why nitrite levels in cured meat are regulated.
An aryl diazonium salt is worth making because its N2+ group leaves so easily. It leaves as nitrogen gas, N2, one of the most stable molecules there is. Releasing a gas also raises the entropy, the disorder, of the system, and the gas bubbles out of the solution, so the loss cannot be reversed. That makes N2+ one of the best leaving groups in organic chemistry.
So the N2+ group can be swapped for many other groups. Copper(I) salts replace it with Cl, Br or CN (the Sandmeyer reactions), KI replaces it with I, and other reagents give F, OH or a plain H. Electrophilic aromatic substitution cannot put three of these groups, OH, CN and F, on a benzene ring at all. It can add iodine only when an oxidant is added along with the I2 (Electrophilic aromatic substitution). The diazonium route needs only KI. Diazonium salts & Sandmeyer works through the whole menu, along with azo coupling, the reaction that built the dye industry.
That page also shows the synthetic payoff: an amino group on a ring can be a temporary tool. Put a nitro group on the ring, reduce it to an amine (Making amines), and let the amine’s strong ortho/para directing effect place the next group (Ortho/meta/para directing effects). Then turn the amine into a diazonium salt and swap it for whatever the target needs, or remove it. A primary aromatic amine is the way into all of this chemistry.
What carries forward
Reductive amination and acylation are the standard ways to put a group on nitrogen in any synthesis. Diazonium chemistry links back to Aromatic Chemistry: it lifts the limits that directing effects put on which substitution patterns a ring can have. And the theme of this page is worth carrying into every later chapter: control a reaction by making its product unreactive.