Nucleophilicity of Amines
Last updated: June 21st, 2019 |
Bonus Topic – Nucleophilicity Of Amines
The relative nucleophilicity of amines doesn’t get a lot of coverage (translation: doesn’t get tested) in many organic chemistry courses, but if we’re going to cover amines, it seems worthwhile to at least devote one post to their nucleophilicity trends.
Most of what follows shouldn’t come as a great surprise, as it will echo a lot of concepts and themes that have made previous appearances in the course (Org 1 in particular). There are, however, a few interesting wrinkles which might help to put some of these concepts into a broader focus.
Table of Contents
- A Quick Review: Basicity vs. Nucleophilicity
- How Acidity/Basicity Is Measured vs. How Nucleophilicity Is Measured
- Some Rules Of Thumb For Comparing Basicity and Nucleophilicity
- A Good First Approximation: “The Stronger The Base, The Stronger The Nucleophile”
- Exception #1: Is This Too Basic For You? A Cautionary Note About Amide Bases
- Exception #2: Bulky Amines
- Why Is Nucleophilicity Much More Sensitive To Steric Effects Than Basicity?
- Exception #3: Poorly Basic Amines That Are Great Nucleophiles
- Nucleophilicity of Amines: Summary
- Bonus Section: Quantifying The Nucleophilicity Trends of Amines With The Mayr Scale
“Um, how are basicity and nucleophilicity different again“? you might ask. Always a good question, because sometimes it takes approaching the subject from a few different angles to really make it sink in. Here, we’ll look at it from the perspective of amines, but a lot of the principles are general.
When Is An Amine Called A “Base” vs. When Is An Amine Called A “Nucleophile”?
- When an amine reacts to form a bond with a proton (H+) we say it is acting as a base. (A Brønsted base, specifically, if you want to tie it back to a familiar concept from general chemistry). The reactive partner of a base is called an “acid”, “Brønsted acid”, or simply a “proton”.
- When an amine reacts to form a bond with any atom other than H, we say it is acting as a nucleophile. The reaction partner is called an electrophile.
Since the electrophile can potentially be any atom on the periodic table (except for H, which would make nitrogen a “base”), there’s inherently a lot more variability possible for nucleophilicity than there is for basicity. Here, we’re going to confine ourselves largely to the reactions of amine bases with carbon-based electrophiles, since it’s the most relevant for our purposes.
- Acid-base reactions are generally reversible and therefore the equilibrium constant (“acidity constant”) Ka can be measured. The (negative) log of the acidity constant, pKa , measures the strength of acids, and it goes from about –10 for strong acids (–10 for HI ) to over 50 for weak acids (>50 for alkanes).
Since “the weaker the acid, the stronger the conjugate base”, a convenient way to compare the basicity of amines is by the pKa of the conjugate acid (which we can call pKaH). The higher the pKaH, the greater the basicity of the amine. (see: Basicity of Amines and pKaH)
- In contrast, nucleophile-electrophile reactions are generally irreversible and therefore equilibrium constants can’t be readily measured. The only option available is to measure the rate.
So what we call “strong” nucleophile is one that reacts quickly with a given electrophile, whereas a “weak” nucleophile reacts slowly with the same electrophile.
The position of an equilibrium (“how stable are two reactants, relative to their products”) and the reaction rate (“how rapidly do two reactants combine to give a product?” ) are two very different measurements. To be sure, there’s a lot of correlation, but it’s just as important to know the situations where the correlations DON’T hold.
To put this in perspective, let’s tie it back to human relationships.
Does the fact that two people fall in love with each other make for a stable long-term marriage?
In fairy tales, the answer is, “of course”. It’s meant to be! What could go wrong?
In real life, the course of love does not always run smooth. That’s why we have country songs.
In general, the stronger the base, the stronger the nucleophile (you can think of this as the “fairy tale” scenario I mentioned above).
Think of this sketch, which shows basicity increasing with nucleophilicity (note: this is not drawn to any reasonable scale)
[Sometimes helpful to think of “basicity” as: “stability of the lone pair”. Anything which makes the lone pair more stable will decrease the basicity. Can’t remember the factors which affect basicity? We covered 5 important basicity trends for amines here. Alternatively you can see this footnote, which covers the effect of conjugation, electron donating (and withdrawing) groups, and hybridization.]
So if basicity mostly correlates with nucleophilicity, you might ask: what are the exceptions?
There are three big exceptions to this rule of thumb: 1) strong bases, 2) bulky amines, and 3) a few highly nucleophilic but weakly basic nitrogens.
The conjugate bases of amines (“amide bases”) such as sodium amide (NaNH2) are extremely strong bases (pKaH about 35–38). If basicity correlates with nucleophilicity, one might expect them to also be great nucleophiles, way up in the right-hand corner of this chart.
In practice, using an amide base for an SN2 reaction is like trying to do surgery with a skill saw. The extra power doesn’t help very much, and there’s too much potential for collateral damage.
Neutral amines are already decent enough nucleophiles for most purposes (e.g. the SN2) and using an amide base doesn’t confer any great advantage.
The potential drawback in using amide bases as nucleophiles is that acid-base reactions tend to be fast, relative to reactions at carbon [Post: Acid Base Reactions Are Fast], and side reactions can occur.
For example, trying to do an SN2 on an alkyl halide may end up with some of the desired product, but it also has the potential to deliver the (undesired) elimination product as well. Too much basicity can be a detriment.
[This is similar to the reason why Grignard reagents aren’t often used in nucleophilic substitution reactions; they’re extremely strong bases, and other processes can get in the way. ]
Pro tip: a better substitute for using NaNH2 in a substitution reaction is to use highly nucleophilic but relatively non-basic sodium azide (NaN3) followed by reduction of the azide to the amine (e.g. with LiAlH4 ).
In acid-base reactions, the electrophile is always the same (H+). But in reactions of nucleophiles, the electrophile can be any number of different atoms on the periodic table, and that can lead to complications.
One way this comes up is that nucleophilicity is much more sensitive to steric effects than is acidity/basicity.
For example, t-butylamine is about as basic as other primary amines but is considerably less nucleophilic.
It comes down to the fact that when an amine is acting as a “base”, it is attacking a fundamentally different type of orbital than when it is acting as a “nucleophile” , and those orbitals have very different steric constraints.
- When an amine base reacts with a proton in an acid-base reaction, the amine lone pair just has to come into contact with the spherical 1s orbital of a hydrogen atom, in order for a reaction to occur.
- When an amine nucleophile attacks an alkyl halide, the nucleophile can’t approach the carbon atom from just any direction in order for a reaction to occur. Remember the “backside attack?” The nucleophile has to approach from a very specific direction so that it can contact the sigma* orbital of a carbon-leaving group bond.
Any additional groups attached to that carbon can further constrain the available angles by which the nucleophile can successfully approach the alkyl halide.
This is the source of the familiar rule of thumb that SN2 reactions are fastest with methyl and primary alkyl halides, and slowest with tertiary alkyl halides.
You can also compare it to the great experiment performed by sportswriter Todd Gallagher when the Washington Capitals repeatedly failed to score on a goalie wearing a fat suit. [See: Steric Hindrance Is Like A Fat Goalie]
So if there are relatively basic amines which are not good nucleophiles, you might ask: are there relatively non-basic amines which are great nucleophiles?
Yes. See the top-left of this graph:
While not technically an amine, the poorly basic (pKaH = 4.7) but highly nucleophilic azide ion (N3 –) immediately comes to mind in this category. Besides being very small and therefore not subject to steric constraints, the azide ion is a “twofer”: each end of the nucleophile can potentially act as a nucleophile. (You can think of this as doubling its effective concentration.)
Two other prominent examples are hydrazine and hydroxylamine, which are both about as basic as ammonia, but are far more nucleophilic. [You may recall that hydrazine is used to cleave phthalimide in the Gabriel synthesis].
What makes these species particularly nucleophilic?
It’s not perfectly understood, but essentially when a nucleophilic atom is bonded to an atom which also bears a lone pair, the nucleophilicity increases significantly [the energy of the HOMO is increases in energy]. This increase in nucleophilicity is called the “alpha-effect“.
For amines (and related species, like azides) the general trend is that nucleophilicity increases with basicity, with a few exceptions:
- bulky bases (like t-butylamine) are less nucleophilic than expected, due to steric factors
- the azide ion and amines bearing an adjacent atom bearing a lone pair (e.g. hydrazine, hydroxylamine) are more nucleophilic than expected due to the “alpha-effect”.
- extremely strong bases (such as NaNH2) can lead to acid-base side reactions rather than the desired reaction with the electrophile.
The usual caveats about nucleophiles apply too. For example, polar aprotic solvents are better than polar protic solvents for promoting SN2 reactions, since they don’t hydrogen-bond with the nucleophile (and thus surround it with a bulky shell of solvent).
If you’re OK with a relatively qualitative approach, great!
On the other hand, if you want a slightly more quantitative approach to nucleophilicity, read on.
An impressive body of work on the relative power of nucleophiles comes from the lab of Herbert Mayr, who for the past few decades has compiled various reactivity scales for a number of different nucleophiles and electrophiles. Given the large number of variables involved, no scale will ever perfect under all conditions, but the values obtained do give us a relative sense of the impact of various factors in the nucleophilicity of amines.
The Mayr numbers are useful as “ballpark” figures for determining order-of-magnitude electronic and steric effects, so long as the electrophile is not greatly sterically hindered.
Like pKa values, the Mayr nucleophilicity parameters are logarithmic. In the case of the Mayr tables, the higher the number, the better the nucleophile.
We won’t go into the details of how these numbers are determined here, but for the curious there’s plenty of background on the website and in the associated papers. Here is Mayr’s homepage database. Here is the the list of amine nucleophilicities that this part of the post is based on. Here is a handy-dandy presentation discussing Mayr’s work.
Quantifying The Effect Of Solvent: Using Polar Aprotic Solvents Is Worth It
We’ve tried to stress how polar aprotic solvents increase nucleophilicity over polar protic solvents since the SN2 days in Org 1. Having a set of nucleophilicity parameters allows us to answer: how much?
The answer appears to be by at least a factor of 100, if we compare the nucleophilicity parameters of ammonia in water and acetonitrile. All else being equal, that’s the difference between waiting 15 minutes for your reaction to complete versus waiting a full day, just by choosing a different solvent. If time is money, what does that make polar aprotic solvents?
The Main Trend: Correlation of Basicity With Nucleophilicity
We mentioned that the nucleophilicity of amines proceeds in the order NH3 < primary amines < secondary amines, and correlated this with pKaH. The Mayr nucleophilicity parameters have the same broad trend.
For water as solvent, the nucleophilicity parameter for NH3 is 9.5, for ethylamine 12.9 (about 1000 times more nucleophilic) and for diethylamine 14.7 (about 100 times more nucleophilic than ethylamine and 100,000 times more nucleophilic than ammonia.)
This starts to put some perspective on the question of why the reaction of ammonia with alkyl halides is not a useful reaction for preparing primary amines; the product is 1000 times more nucleophilic!
Another example where basicity correlates with nucleophilicity is in the effect of electron-withdrawing groups. Compare the nucleophilicity of piperidine (18.1 in H2O) with morpholine (15.6 in H2O). The electron-withdrawing oxygen has the effect of reducing nucleophilicity by a factor of 300.
With 2,2,2-trifluoroethylamine, the trifluoro group reduces nucleophilicity by about 100,000.
Quantifying The Effect Of Steric Hindrance
We mentioned that steric hindrance reduces the nucleophilicity of amines. By how much?
The Mayr parameters for t-butylamine, isopropylamine and n-propylamine show a clear trend, going from (in water as solvent) 10.5 (in water) for t-butylamine up to 12.0 for isopropylamine to 13.3 for n-propylamine.
So as a rough gauge, the t-butyl group reduces the nucleophilicity by a factor of about 1000 versus a “normal” primary amine.
Highly nucleophilic amines that aren’t very basic
The measured set of nucleophilicity parameters give an estimate that hydroxylamine is about 100 times more nucleophilic than ammonia (in water), while hydrazine is about 10,000 times more nucleophilic.
The nucleophilicity parameter for the azide ion wasn’t measured in water, but if we can compare two different polar aprotic solvents (DMSO and acetonitrile) it’s about a billion times more nucleophilic than NH3.
Useful tidbit: never, ever use CH2Cl2 or chloroform as solvent for any reaction involving the azide ion. The azide ion is so nucleophilic it will displace the chlorides, leading to the formation of a potentially explosive diazidomethane.
What About Tertiary Amines?
Conspicuously absent from this whole discussion has been the question of tertiary amines.
- On one hand, tertiary amines have an extra alkyl group attached to nitrogen, so we’d expect them to be more basic (and nucleophilic) than secondary amines. [In fact, triethylamine is slightly less basic than diethylamine in water, largely because of solubility factors].
- On the other hand, tertiary amines should be more sterically hindered than secondary amines, reducing nucleophilicity.
So are tertiary amines better or worse nucleophiles than secondary amines? Which factor “wins” ?
The Mayr nucleophilicity factors can help sort out this question.
It’s hard to see trends with water as a solvent because most tertiary amines are not very soluble in water.
In acetonitrile, it looks like tertiary amines are at least one, if not two orders of magnitude more nucleophilic than comparable secondary amines.
A particularly interesting example of a highly nucleophilic tertiary amine is quinuclidine, where the three alkyl groups are “tied back” in a bicyclic structure. The nucleophilicity paramter of quninuclidine in acetonitrile is 20.5, on par with the azide ion.
A caveat: Remember that the Mayr parameters work best with non sterically hindered electrophiles. So expect the reaction rate of a tertiary amine with a hindered electrophile to fall off a cliff, relative to a less hindered electrophile.
- Basicity (and nucleophilicity) is reduced by conjugation, when a lone pair can be delocalized through resonance (e.g. aminobenzene is less basic than cyclohexylamine, and less nucleophilic also.)
- Electron withdrawing groups reduce basicity (and nucleophilicity), such as in morpholine versus piperidine (sigma acceptors) and acetamide (pi acceptor) versus ethylamine.
- Conversely, electron donating groups increase basicity (and nucleophilicity) such as in alkylamines and 4-dimethylaminopyridine vs. pyridine (pi donors).
- Increasing the s-character of the orbital decreases basicity (and nucleophilicity) sp < sp2 < sp3
In this post we’re mainly discussing nitrogen nucleophiles and carbon electrophiles, but when dealing more broadly on the subject of nucleophilicity the issue of orbital overlap between orbitals on different rows of the periodic table can arise. Sometime in the future we can address hard soft acid base (HSAB) theory. On the other hand, at a lecture I saw Mayr give once on his reactivity tables, he was asked where hard and soft acid-base theory effects were in his data. His answer was, “I don’t know”, by which he meant, he thought his data doesn’t seem to support it.