The azide ion (N3– ) is one of those chemical entities that just plain looks weird.
- Three nitrogens in a row? check
- Two negative charges and a positive charge, for a net charge of –1 ? check
- Stable? check (as long as it’s treated gently).
As it turns out, the azide ion can extremely useful for forming C-N bonds in nucleophilic substitution reactions.
The Azide Ion Is A Great Nucleophile
The azide ion is the conjugate base of hydrazoic acid, HN3. Despite being only weakly basic (the pKa of HN3 is only 4.6) N3 is an extremely good nucleophile – according to one measure, more nucleophilic than any amine (see post: Nucleophilicity of amines). It’s not hard to rationalize why this might be when you stop to think about it. With four nucleophilic lone pairs confined to a very small volume, the likelihood of a collision with an electrophile that results in a reaction is much higher than it would be for an amine with bulky alkyl groups, to take just one example.
You can think of N3– as a lean, mean, nitrogen-tipped missile that delivers its payload onto carbon quickly, efficiently, and without significant side reactions.
In SN2 reactions, primary and secondary alkyl halides and sulfonates are readily displaced by N3– , resulting in alkyl azides:
The usual procedure is to use an azide salt such as NaN3 or KN3 with the appropriate alkyl halide in a polar aprotic solvent such as acetonitrile (CH3CN) or dimethylsulfoxide (DMSO).
The organic azide products are reasonably stable, even if they look a little weird. Some have even found use as pharmaceuticals (e.g. AZT, below right) or as useful probes for studying chemical biology. [More in footnotes].
For our purposes the most useful property of organic azides is that they serve as “masked” amines.
Azides as “Masked Amines”
Note the resonance form above with a nitrogen-nitrogen triple bond. If treated with a reducing agent, such as LiAlH4 or even catalytic hydrogenation (Pd/C , H2) organic azides can be reduced to primary amines, liberating N2 in the process.
This makes for a very useful route to primary amines from alkyl halides!
We’ve previously explored the Gabriel synthesis as a route to primary amines [see post: The Gabriel Synthesis], but this route is superior because the azide can be reduced to the amine under gentle conditions (e.g. hydrogenation). None of that high-temperature cleavage of the phthalimide with hydrazine to worry about.
We also saw that making primary amines through direct treatment of alkyl halides with NH3 often doesn’t result in the desired product, because of amines have a Cookie-Monster-like tendency to react multiple times with alkyl halides: since the amine products tend to be more nucleophilic than the reactants, it’s hard to get amines to stop at munching just one alkyl halide. [see post: Alkylation of Amines]
[footnote – primary amines do’s and don’ts graphic]
Exploding On The Launchpad
If N3 is a “nitrogen-tipped missile”, as we commented above, it’s worth recalling what happens when a missile isn’t handled with care: it’s essentially a bomb.
NaN3 and KN3 are white powders that can be stored indefinitely at room temperature and scooped out at the bench without any trouble.
But when heated or subjected to shock – all bets are off. The same goes for exposure of these azide salts to acid, which forms potentially explosive HN3.
And NaN3 and KN3 are the nice azides! If mixed with metal salts of lead, mercury, cadmium, zinc, or silver, even more explosive azide compounds can result. These metal azides are contact explosives, some of which are so sensitive that they will detonate if someone farts twenty feet down the hall. Some crazy chemists even make these things on purpose, but we will not be travelling anywhere near the blast radius of that topic on this blog anytime, ever.
NaN3 and KN3 should be used at reasonably dilute concentrations, behind a blast shield on preparative scale, and incidentally never used with CH2Cl2 solvent, which can result in highly explosive diazidomethane.
Now, there are times when a sudden blast of nitrogen gas is useful – and potentially life-saving. You may have already spent a portion of your day in close proximity to sodium azide without even realizing it. That’s because NaN3 is one of several propellants that finds use in airbags; once triggered by an accelerometer, detonation of NaN3 expels N2 gas at over 200 mph, inflating the bag in 1/25 of a second.
Yay for azide salts!
(But don’t be this guy):
Bottom line: SN2 reactions between alkyl halides or sulfonates with azides are probably the single best way to synthesize primary amines from alkyl halides. It certainly beats the Gabriel synthesis.
Nucleophilic Acyl Substitution And The Curtius Rearrangement
The SN2 reaction isn’t the only type of substitution reaction we’ve explored. There’s also nucleophilic acyl substitution. A carbonyl attached to a good leaving group (such as an acid chloride or anhydride) will undergo substitution when an appropriate nucleophile is added.
In the first step of the mechanism, the nucleophile (Nu) attacks the carbonyl carbon, forming C-Nu and breaking C-O (pi). This gives rise to a tetrahedral intermediate. In the second step, the C-O pi bond is re-formed, and the carbon-(leaving group) bond is broken, resulting in the nucleophilic acyl substitution product.
Here is an example of a nucleophilic acyl substitution reaction between an acid chloride and the N3 ion to give an acyl azide:
The most common use for acyl azides is that upon heating, they rearrange to give isocyanates in a reaction known as the Curtius rearrangement. [see post: The Hofmann and Curtius Rearrangements]
- If the Curtius is performed in the presence of an alcohol such as methanol, a carbamate is formed.
- If water is added instead, an unstable carbamic acid is briefly formed, which then loses carbon dioxide to give a primary amine.
Why might this be useful? Here’s one way.
There are a lot of methods for making aromatic carboxylic acids, but not so many great ways of forming C-N bonds on aromatic rings. Say you have a benzoic acid (such as p-methylbenzoic acid) but need to form a (protected) aromatic amine. The Curtius protocol represents a nice way of forming a new C-N bond on the ring without having to resort to nitration + reduction.
(In this way it’s somewhat analogous to an nitrogen equivalent of the Baeyer-Villiger oxidation. )
More examples of the azide ion as nucleophiles
Two more applications of the azide ion as a nucleophile include the opening epoxides (in what is essentially an SN2 process – note the inversion of configuration) and also conjugate addition on alpha,beta unsaturated ketones:
The resulting alkyl azides can then be reduced to primary amines using the methods covered above.
An interesting and useful application of organic azides that hasn’t yet made it into introductory organic chemistry textbooks (but likely will in the near future) is a process technically called the “copper-catalyzed azide-alkyne cycloaddition“, but more commonly just called the “alkyne-azide click reaction”.
When a terminal alkyne is treated with a copper catalyst in the presence of an organic azide, a cycloaddition rapidly results, leading to the formation of a 1,2,3-triazole.
We won’t go deep into the details, but this ring-forming reaction involves 6 pi electrons and is a cousin of the Diels-Alder reaction (nitpckers will rightly note that the copper-catalyzed version isn’t technically a concerted process, but the non-catalyzed version is)
The reaction is notable for a few reasons:
- the reactive partners (terminal alkyne and organic azide) are relatively inert in the absence of a catalyst under most conditions.
- once copper is added, reaction occurs extremely rapidly under mild conditions and in a variety of solvents (including water).
- it’s extremely selective under these conditions,the alkyne and azide partners react only with each other, irreversibly.
- the product, a triazole ring, is aromatic and stable
It’s been called a “click” reaction by one of its originators, Barry Sharpless, because it’s one of the closest things chemists have to a reaction that is like the smooth “clicking together” of two complimentary lego blocks.
What has made this reaction particularly useful of late is that azide and alkyne partners can be installed in molecules of biological interest, and “clicked” together when desired.
- Putting an acetylene at the 5′ position of a sugar and an azide at the 3′ position has been used to make “clicked” DNA, where the triazole replaces the organophosphate. [link]
- Incredibly, one group has made “clicked” DNA, put it into human cells, and managed to get it transcribed by messenger RNA. [link]
- azide-containing AZT (above) works by gumming up the transcription machinery. Some clever folks found a way to apply the azide-alkyne click reaction toward rapid genome sequencing, a technique known as ClickSeq. [a very readable account here, courtesy of In The Pipeline]
Ammonium nitrate was about one-tenth the price of tetrazole, according to Upham, who also reviewed industry patents. But ammonium nitrate had a critical flaw that he says led other air bag makers to give up on it: Ammonium nitrate has five phases of varying density that make it hard to keep stable over time. A propellant made with ammonium nitrate would swell and shrink with temperature changes, and eventually the tablet would break down into powder. Water and humidity would speed the process. Powder burns more quickly than a tablet, so an air bag whose propellant had crumbled would be likely to deploy too aggressively. The controlled explosion would be just an explosion. “Everybody went down a certain road, and only Takata went down another road,” says Jochen Siebert, who’s followed the air bag industry since the 1990s and is now managing director of JSC Automotive Consulting. “If you read the conference papers from back then, you can actually see that people said, ‘No, you shouldn’t. It’s dangerous.’ ”
Shrapnel from the faulty airbags has killed 13 people worldwide, and injured more than 100.
On June 25, 2017, facing liabilities of over $11 billion dollars, Takata filed for bankruptcy.