Dienes and MO Theory
Regiochemistry In The Diels-Alder Reaction
Last updated: October 27th, 2022 |
Regiochemistry (“Regioselectivity”) In The Diels-Alder Reaction
The Diels-Alder is an onion, and we just keep peeling back the layers.
- When non-symmetrical dienes react with non-symmetrical dienophiles, two regioisomers (constitutional isomers) are possible.
- Dienes with substituents on the terminus (“1-substituted dienes”) tend to give “1,2” products (nicknamed “ortho”).
- Dienes with substituents on the 2-position (“2-substituted dienes”) tend to give the “1,4” product (nicknamed “para”).
- In general, “1,3” products (“meta”) are only minor byproducts.
Table of Contents
- Regiochemistry In The Diels-Alder Reaction
- Flashback: Markovnikov’s Rule and “Regioselectivity”
- The Diels-Alder Reaction Is Regioselective
- Summary: Regioselectivity In The Diels-Alder Reaction
- Notes (more on the origin of regioselectivity)
- (Advanced) References and Further Reading
1. “Regiochemistry” In The Diels-Alder Reaction
So far in the Diels-Alder, we’ve seen examples of:
- symmetrical dienes with symmetrical dienophiles
- unsymmetrical dienes with symmetrical dienophiles
- symmetrical dienes with unsymmetrical dienophiles
These three situations (laid out in the image below) each have the potential to form stereoisomers (i.e. diastereomers and/or enantiomers). But all the products have exactly the same connectivity.
This brings us to a particularly challenging case. What happens when we have an unsymmetrical diene reacting with an unsymmetrical dienophile?
Why is this situation different from the first three?
Because there are two different ways for the diene and dienophile to “line up! “. In the example above, the diene and dienophile can come together two ways:
- “head to head” such that the methyl group on the diene and the C=O bond point in the same direction (giving the top product);
- “head to tail” such that they point in the opposite direction (bottom product).
How are these products related to each other, overall (not counting stereochemistry)?
Since they have the same molecular formula but different connectivity, they’re constitutional isomers.
Now comes the big question. In this kind of a Diels-Alder, are these products formed in roughly equal ratios… or is there a preference for one type of connectivity over another?
The short answer is, “yes, the Diels-Alder has a preference for one type of connectivity.”
In other words, the reaction has regioselectivity.
Wait. What’s regioselectivity? Quick review:
2. Flashback: Markovnikov’s Rule and “Regioselectivity”
Where have we seen this type of situation before?
With alkenes! We saw that adding an acid like HCl to an alkene like 1-butene, we could obtain two possible products: 1-chlorobutane or 2-chlorobutane. [See: Markovnikov’s Rule]
These products have the same molecular formula, but different connectivity, which makes them constitutional isomers.
However, we saw that the reaction doesn’t give an equal ratio of products. Instead, there’s about a 4:1 preference for 2-chlorobutane over 1-chlorobutane, which we saw was due to a greater preference for the more stable carbocation intermediate.
This preference goes by the name “Markovnikov’s Rule”:
The tendency of alkenes to obey “Markovnikov’s rule” in these reactions is an example of regioselectivity. The reaction is selective in forming one constitutional isomer over another.
(We say “selective” and not “specific” because more than one product is formed. Use of the prefix “regio” comes from the observation that the chloride tends to attack one region of the double bond, and the proton, another. )
It’s worth noting that 2-chlorobutane is formed as a racemic mixture of enantiomers. So while the reaction is regioselective, it is not enantioselective.
3. The Diels Alder Is Regioselective
Like the reaction of acids with alkenes, the Diels-Alder reaction is regioselective. Two main cases will illustrate the point.
Case 1: 1-substituted dienes
The first important case concerns a diene with a substituent on the “1” position of the diene, such as 1-methoxybutadiene. (I know that if a methyl group were present instead, it would technically be the “4” position according to IUPAC – bear with me on this).
Consider the Diels-Alder of 1-methoxy butadiene with methyl methacrylate. There are two ways that the diene and dienophile can connect.
- Line it up one way (“head to head“, below), and you get a new six-membered ring where two substituents are on two adjacent carbons (a “1,2” relationship)
- Line it up another way (“head to tail“), and the two substituents have a “1,3” relationship.
These two products have different connectivity and are therefore constitutional isomers (“regioisomers”).
By analogy to aromatic nomenclature, the “1,2” and “1,3” patterns are nicknamed ortho- and meta- respectively. [I say “nicknamed” because these are terms of convenience, nothing else. Don’t tell IUPAC!]
So which of the two Diels-Alder products is favored? The “ortho” or the “meta” ?
Here’s what’s experiments tell us:
The ortho product is major and the meta product is minor.
This holds for a large number of 1-substituted dienes; I’ll just show two. Hans Reich at UW-Madison has a longer list of examples – see here.
- With 1-methoxybutadiene, the ortho is the only product!!
- When a methyl group is in that position, the ortho outnumbers meta by about 8:1
Case 2: 2-substituted dienes
The second important case is when there’s a substituent on the 2-position of the diene, such as 2-methylbutadiene.
Again, there are two ways it can line up, except this time it’s to provide para (1,4) and meta (1,3) products.
So which of these products is dominant?
The para product is major and the meta product is minor.
- 2-methoxybutadiene favors the para product by about 8:1
- 2-methylbutadiene favors the para product by about 2:1
What if there’s substituents on both the 1- and 2- positions? In these cases, it turns out that the substituent on the 1-position is more powerful at directing the products. [See Note 1]
4. Summary: Regiochemistry In The Diels-Alder Reaction
Avoid the meta- product (1,3). That’s really it.
So why does the Diels-Alder wind up this way?
For the answer, read on…..
It can be helpful to think of the diene as a nucleophile and the dienophile as the electrophile.
The dominant product will be the one where the most nucleophilic carbon on the diene lines up with the most electrophilic carbon on the dienophile.
So how do we determine what these carbons are?
- Look for the second-best resonance form of the diene and the dienophile! (sometimes known as Grossman’s rule)
- Line up the negative charge on the diene with the positive charge on the dienophile.
This will get you to the right result!
Note how this results in the ortho– product and not the meta- product, just like what’s observed in experiment.
What about 2-substituted dienes?
Same process. Draw out the “second-best” resonance form. Now line up the negative charge on the nucleophile with the positive charge on the “2nd best” resonance form of the electrophile:
Note how this correctly predicts the para product will be favored over the meta product.
Now… is this really the best way to do it?
For our purposes, yes.
For more advanced purposes… we rely on molecular orbital calculations. In advanced courses, we talk a lot about the size of “coefficients” on the HOMO or LUMO of the diene/dienophile, and we’re not going there. If you are interested in this topic, I highly recommend Ian Fleming’s Frontier Orbitals and Organic Chemical Reactions. Classic book.
What About 1,2-Substituted Dienes?
Note 1. What if they’re both substituted? Here’s a practical example.
What’s observed is that the 1-position on the diene has a greater influence on the product than the 2-position.
(Advanced) References and Further Reading
- Frontier molecular orbital theory of cycloaddition reactions
Kendall N. Houk
Accounts of Chemical Research 1975, 8 (11), 361-369
- Generalized frontier orbitals of alkenes and dienes. Regioselectivity in Diels-Alder reactions
K. N. Houk
Journal of the American Chemical Society 1973, 95 (12), 4092-4094
- Modeling chemical reactivity. 1. Regioselectivity of Diels-Alder cycloadditions of electron-rich dienes with electron-deficient dienophiles
D. Kahn, C. F. Pau, L. E. Overman, and Warren J. Hehre
Journal of the American Chemical Society 1986, 108 (23), 7381-7396
- Quantitative Characterization of the Local Electrophilicity of Organic Molecules. Understanding the Regioselectivity on Diels−Alder Reactions
Luis R. Domingo, M. José Aurell, Patricia Pérez, and Renato Contreras
The Journal of Physical Chemistry A 2002, 106 (29), 6871-6875
- The regioselectivity of the diels-alder reaction between a diene with an electron-donating substituent and a dienophile with an electron-donating substituent: a test case for frontier orbital theory
Ian Fleming, Federico L. Gianni, Talat Mah
Tetrahedron Lett. 1976, 17 (11), 881-884
Better than the Ian Fleming of James Bond fame – but his book on Frontier Molecular Orbital theory is excellent and recommended for all advanced students of organic chemistry.
- Regioselectivity in Hetero Diels–Alder Reactions
Carla Grosso, Marta Liber, Amadeu F. Brigas, Teresa M. V. D. Pinho e Melo, and Américo Lemos
Journal of Chemical Education 2019, 96 (1), 148-152
- On the Brassard’s rule of regioselectivity in Diels–Alder reactions between haloquinones and polar dienes
Mauricio Maldonado-Domínguez, Karen Ruiz-Pérez, Oscar González-Antonio, Margarita Romero-Ávila, José Méndez-Stivalet and Blas Flores-Pérez
RSC Adv., 2016, 6, 75194-75201
- Origins of Regioselectivity of Diels−Alder Reactions for the Synthesis of Bisanthraquinone Antibiotic BE-43472B
Amy E. Hayden, Robert S. Paton, Jochen Becker, Yee Hwee Lim, K. C. Nicolaou, and K. N. Houk
The Journal of Organic Chemistry 2010, 75 (3), 922-928
Experimental and theoretical study on a particularly complex diene/dienophile pairing.
7 thoughts on “Regiochemistry In The Diels-Alder Reaction”
Is there an answer to the question by JM yet?
Hi James, this website has been a tremendous help for actually understanding (and not just regurgitating) organic chemistry, so thank you so much for that! I just noticed that, in the “Case 1: 1-substituted dienes” section, the fourth dot point mentions an 8:1 ratio between ortho and para, however the subsequent image shows an 8:1 ratio between ortho and meta. Is there also an ortho:para relationship here, or is the dot point supposed to be ortho:meta? Thank you again!
It *is* supposed to be ortho: meta. Thanks for the spot! Fixed
First of all this entire website has been so helpful, thank you so much for helping clarify all these confusing concepts!
In your example of the 1,2 disubstituted diene you have 2 identical substituents. What would happen in the reaction if you had a 1,2 disubstituted diene with different substituents (say, for example, a methyl on position 1 and methoxy on position 2)? Is one substituent more “powerful” than the other and therefore more influential in determining the final Diels-Alder product, even though the position 1 substituent is the one that determines the final structure like you said? I read online that methoxy is a more powerful directing group than methyl. Not sure how true that is or if it’s relevant to this. Thank you!
That is a great question. Methoxy is a more powerful directing group than methyl, but directing groups on the 1-position are more dominant than directing groups on the 2-position. I will have to look it up. Thanks for asking.
Hi, quick question. In regards to ‘1-substituted dienes’ and ‘2-substituted dienes’ can electron withdrawing groups vs electron donating groups affect the ‘head:head or head:tail’ alignments and therefore break the given rule that a ‘1-substituted diene’ always gives ortho product?
Good question! The result will still be “ortho” and “para” products. If you start adding electron withdrawing groups to the diene, then it will tend to undergo inverse electron-demand diels alder reactions (i.e reacting with electron rich “dienophiles”). So something like 2-cyanobutadiene will react with ethyl vinyl ether, but still give para, not meta!