Reactions of Aromatic Molecules
Activating and Deactivating Groups In Electrophilic Aromatic Substitution
Last updated: March 11th, 2020 |
Quick summary on what we’ll talk about today, on activating and deactivating groups in electrophilic aromatic substitution:
There’s a lot to this post, so here’s a quick index:
Table of Contents
- Activating And Deactivating Groups
- Measuring Reaction Rates Can Provide Insight Into The Mechanism
- “Activating” and “Deactivating” Groups – A Definition
- “Sigma” (σ) donors and acceptors (otherwise known as “inductive effects”)
- Pi ( π) Donors and Acceptors (often just called “Resonance”)
- Oxygen And Nitrogens Containing Lone Pairs Are Highly Activating When Bonded Directly To The Ring
- Halogens (F, Cl, Br, I) Are Deactivating
- Pi Acceptor Groups Are Strongly Deactivating
- A Table of Activating and Deactivating Groups
- Summary: What Does This Tell Us About The Mechanism Of Electrophilic Aromatic Substitution?
Last post in this series we introduced electrophilic aromatic substitution. Here’s the general case:
Why is this a substitution reaction, you ask? Because we’re forming and breaking a bond on the same carbon. We form C–E (where “E” is a generic term for “electrophilic atom”) and we break C–H.
[As for the specific identity of “E”, we mentioned six key electrophilic aromatic substitution reactions in the last post (bromination, chlorination, nitration, sulfonylation, Friedel-Crafts alkylation and Friedel-Crafts acylation) that we’ll eventually dig into in detail. But not yet. ]
So if that’s the summary of what happens, the next obvious question is: how does it happen?
In other words, what’s the mechanism?
Obligatory pre-mechanism speech: You can’t determine the mechanism of a chemical reaction merely through logical deduction from first principles. Sure, you can make guesses – even good ones! But the ultimate test of a mechanistic hypothesis is how well it fits with experiment, and that typically involves a lot of lab work. What you’re taught in an introductory course is the tippy-topmost layer of snow on the iceberg. We give you the best answer, and in retrospect it looks obvious. What you don’t see is all the failure, wrong turns, and false hypotheses that happened along the path towards determining the correct mechanism. However, the mechanisms of these reactions that you will learn about weren’t obvious to most of their discoverers, who were among the brightest and best chemists of their time. Remember that.
As far as determining mechanisms is concerned, one of the best tools we have in our experimental arsenal is the ability to measure reaction rates.
By measuring the effect that slight tweaks in the experimental conditions (e.g. structure of reactant, temperature, solvent) have upon the rate, we can gather useful insights about how a reaction operates “under the hood”.
Of the parameters mentioned above, changing the substrate (reactant) is probably the most powerful way to probe a mechanism, because it allows you to tune how electron-rich (nucleophilic) or electron-poor (electrophilic) it is.
Let me show you what I mean.
Let’s arbitrarily pick one electrophilic aromatic substitution reaction: nitration.
- We know that by adding nitric acid and H2SO4, benzene can undergo nitration to form nitrobenzene (break C-H, form C-NO2)
- We can even measure the rate of this reaction at a given temperature, concentration, and solvent.
- Using the exact same experimental conditions we can then measure the rate of the reaction when toluene (methylbenzene, C6H5CH3) is used as the substrate instead of benzene.
- The nitration of toluene is 23 times faster than it is for benzene. [Ref 1]
- Using the exact same experimental conditions, we can also use trifluoromethylbenzene (C6H5CF3) as the substrate, and measure the reaction rate.
- The nitration of trifluoromethylbenzene is 40,000 times slower than it is for benzene (2.5 × 10-5).
This pattern turns out to be general for other electrophilic aromatic substitution reactions as well (chlorination, bromination, Friedel-Crafts, and others).
Let’s call a group activating that increases the rate of an electrophilic aromatic substitution reaction, relative to hydrogen. As we just saw, CH3 is a perfect example of an activating group; when we substitute a hydrogen on benzene for CH3, the rate of nitration is increased.
A deactivating group, on the other hand, decreases the rate of an electrophilic aromatic substitution reaction, relative to hydrogen. The trifluoromethyl group, CF3 , drastically decreases the rate of nitration when substituted for a hydrogen on benzene.
This definition is ultimately based on experimental reaction rate data. It doesn’t tell us why each group accelerates or decreases the rate. “Activating” and “deactivating” just refers to the effect of each substituent on the rate, relative to H.
OK then. So why might CH3 increase the rate of reaction, and CF3 decrease it?
Let’s quickly think back to what we know about alkyl groups (such as CH3) and haloalkyl groups (such as CF3), and try to address this question.
In CH3, the carbon atom is more electronegative (2.5) than hydrogen (2.2). This means that the carbon attracts a bit more than an equal share of electron-density from the covalent bond with H, resulting in a partial negative charge (δ–) on carbon and a partial positive charge (δ+) on hydrogen. This partial negative charge is then available to be donated to an adjacent atom. Hence, we tend to think of CH3 as an electron-rich species; an electron–donor.
In CF3 the electrons are pulled in the opposite direction. Three highly electronegative (4.0) fluorine atoms pull electron density away from the carbon atom (2.5), resulting in a partial positive charge (δ+) on carbon. Rather than donate electron density, the carbon tends to accept (pull away) electron density from adjacent atoms (this is the familiar inductive effect) We generally consider CF3 to be an electron-poor species; an electron-acceptor.
Since these inductive effects operate solely through single bonds (“sigma”, or σ bonds) this behaviour is sometimes called “sigma donation” (as for CH3) or “sigma accepting” (for CF3).
So it seems like a good hypothesis that
- activating groups are electron-donating (relative to H), and
- deactivating groups are electron-withdrawing (relative to H)
Sigma donation and acceptance helps us to understand the effect of alkyl groups on electrophilic aromatic substitution. So what about other functional groups? What effect might, say, a hydroxyl group have on the rate of nitration?
Quiz time. Do you think –OH would be activating (increase the rate) or deactivating (decrease the rate) for electrophilic aromatic substitution (such as nitration)? Guessing is OK!
Based on what we just said, it’s fully understandable if you said, “deactivating”. After all, oxygen is highly electronegative (3.4) and through induction, pulls away electron density through the bond. In other words, it’s a sigma-acceptor.
The fact is, however, that OH greatly accelerates the rate, orders of magnitude more than CH3 does. In fact I couldn’t find good rate data comparing OH to CH3 because in the case of -OH, the reaction is what’s called, “diffusion controlled”. That roughly means, “as soon as the reactant comes in contact with the electrophile, a reaction occurs.” In other words, the –OH group is highly activating.
Clearly, something else must be going on here besides the inductive effect of oxygen!
6. Oxygen And Nitrogens Containing Lone Pairs Are Highly Activating When Bonded Directly To The Ring
As we saw in our chapter way back on resonance, hydroxyl groups are excellent pi donors. The lone pairs on the oxygen atom can form a pi bond with an adjacent atom containing an available p-orbital.
This donation effect (or “resonance”) must outweigh electron-withdrawal via inductive effects, otherwise we’d observe that hydroxyl groups are deactivating.
The same is true for nitrogen groups with lone pairs, such as amines and amides (below).
[One measure of the importance of pi-donation in the activating nature of amines is seen in their behavior under strongly acidic conditions. If the nitrogen lone pair is either protonated with strong acid or undergoes a substitution reaction to form NR3+ , pi-donation is impossible and the group becomes strongly deactivating (see table below). ]
Not all groups capable of pi donation are activating groups. For example, halogens (F, Cl, Br, I) tend to be deactivating. The rates of electrophilic aromatic substitution reactions on fluorobenzene, chlorobenzene, bromobenzene, and iodobenzene are all slower than they are for benzene itself. In these cases, inductive effects (“sigma accepting”) would appear to have a greater effect on the rate than any pi-donation from the lone pairs. [pi donation < sigma acceptance]. [Why?]
A good rule of thumb for pi-donation ability is the basicity of the lone pair. Amines tend to be better bases than oxygens, which are far better bases than halogens.
Alright. What if electrons flow in the opposite direction? Is there an opposite of “pi donor” ?
Yes! As you may already know, the opposite of a “pi-donor” is a “pi acceptor”. Certain functional groups can accept, rather than donate, a pi bond from the ring, resulting in a new lone pair on a substituent atom. Examples are NO2, carbonyl groups (C=O), sulfonyl, cyano (CN) among others. These groups are universally deactivating, slowing the rate of electrophilic aromatic substitution.
In terms of resonance, one can draw a pi bond from the aromatic ring forming a pi bond with the atom bound to the ring, resulting in formation of a new lone pair on an electronegative atom on the substituent. Note how this results in a positive charge on the ring!
So how do we keep all of these factors straight?
This is an example of why I say that resonance is the most important key concept to review for Org 2. In the section on aromatic chemistry it comes back with a vengeance.
Now seems like the right time to present a big table of activating and deactivating groups. It’s hard to rank exactly by power since the effect is averaged over several types of reactions.
I would suggest five main “buckets”, below:
- Nitrogen and oxygens with lone pairs – amines (NH2, NHR, NR2), phenol (OH) and its conjugate base O– are very strong activating groups due to pi-donation (resonance). Alkoxy, amide, ester groups less strongly activating.
- Alkyl Groups – (with no electron withdrawing groups). Moderately activating through inductive effect.
- Halogens – Moderately deactivating. Electron withdrawing (highly electronegative) nature outweighs donation of electron density through a lone pair.
- Atoms with pi-bonds to electronegative groups – Strongly deactivating. NO2, CN, SO3H, CHO, COR, COOH, COOR, CONH2. All pi-acceptors.
- Electron withdrawing groups with no pi bonds or lone pairs – Strongly deactivating. CF3, CCl3, and NR3(+). Pure inductive effect.
Once you remember the somewhat counterintuitive fact that O and N-bonded functional groups with lone pairs are activating, and halogens are deactivating, the rest is fairly straightforward.
One final word. Our table of “activating” and “deactivating” groups turns out to be a little bit like a pKa table. How? We can evaluate several factors that have an impact on pKa, but the ultimate test of which factor is more important is experimental measurement of an equilibrium constant. Likewise, with activating and deactivating groups, we can identify factors which may or may not make a certain group activating or deactivating, but in the end, its position on the chart comes down to experimental measurements of reaction rates.
OK. So what does all of this tell us?
Since the rate is so sensitive to whether the group is electron donating or electron withdrawing (“electronic effects”, as organic chemists might quickly summarize it) it would suggest that the rate determining step is the formation of a fairly unstable electron-poor species, such as a carbocation.
Recall CH3 and CF3. You may recall that the order of carbocation stability (tertiary > secondary > primary) is due to the fact that carbocations are stabilized by adjacent alkyl groups (such as CH3), and, conversely, are destabilized by adjacent electron withdrawing groups (like CF3).
Likewise, carbocations are stabilized by adjacent atoms that can donate lone pairs (e.g. O and N) through resonance, and destabilized by pi acceptors such as C=O, NO2, and so on.
A likely first step would be something like this:
We’ll go into the full mechanism of electrophilic aromatic substitution in the next post, but will fill in additional detail in a bonus topic below.
Two other relevant points for determining the rate of electrophilic aromatic substitution.
1. [Advanced] No deuterium isotope effect is observed in electrophilic aromatic substitution
In electrophilic aromatic substitution a C-H bond is broken. One way to probe the mechanisms of reactions that involve C-H bond cleavage is to use deuterium (D) labelling. In reactions where C-H bond breakage is a rate-determining step (e.g. E2 elimination) a C-H bond can break up to 6-7 times faster than a C-D bond. This is called a deuterium isotope effect and it is measurable.
Electrophilic aromatic substitution reactions have no significant deuterium isotope effects. [Note] This strongly suggests that C-H bond breakage is not the rate determining step.
2. Carbocation intermediates have been isolated that strongly support the proposed mechanism
Here’s a species that’s been observed when 1,3,5-trimethylbenzene (mesitylene) is treated with ethyl fluoride and boron trifluoride at –80°C (this is a Friedel-Crafts alkylation reaction, by the way).
The carbocation intermediate (called an “arenium ion” or “Wheland intermediate” was isolated as a white solid with melting point –15°C, and analyzed by NMR spectroscopy.
As Eric Jacobsen might say: “mechanisms can never be proven, but….” . (this pretty much seals the deal). We’ll go into in more detail in the next post.
Two. Why? Interestingly, fluorine is the most activating of the halogens. The reason is likely that the overlap of the lone pair in the fluorine 2p orbital with the p orbital on carbon is much better (resulting in a stronger pi-bond) than is donation with the 3p (and higher) p orbitals of chlorine, bromine, and iodine.
Actually a white lie; some electrophilic aromatic substitution reactions do have very small deuterium isotope effects, but we’re not touching that topic, nosiree. [partitioning effects, see March’s Advanced Organic Chemistry, 5th ed., p. 679]