11. Pericyclic Reactions – Advanced Organic Chemistry

11

Pericyclic Reactions

LEARNING OBJECTIVES

By the end of this chapter you should be familiar with

  • Types and principles of pericyclic reactions.
  • What determines whether these pericyclic reactions go forwards or backwards.
  • Conservation of orbital symmetry and what conrotatory and disrotatory mean.
  • Electrocyclic reactions and rules governing them.
  • Cycloaddition reactions and rules governing them.
  • Sigmatropic reactions and rules governing them.
11.1 INTRODUCTION

Organic reactions can be classified into three major classes: polar reactions (reaction between nucleophile and electrophile), radical reactions (reaction using one electron from each of the reactants) and pericyclic reactions. A pericyclic reaction, that is, cyclic reorganization of electrons, is one in which bonds are made or broken in a concerted cyclic transition state. A concerted reaction is one that involves no intermediates during the course of the reaction. Pericyclic reactions have certain characteristic properties, although as usual it is not difficult to find exceptions to all these rules.

  • There is little solvent effect on the rate of pericyclic reactions that can occur in the gas phase with no solvent.
  • There is no nucleophilic or electrophilic component. This means that in the arrow-pushing sense, there is no beginning and no ending for the arrows, and the arrow-pushing can occur in either a clockwise or anti-clockwise direction.
  • Normally, no catalyst is needed to promote the reactions. However, many transition metal complexes can catalyze pericyclic reactions by virtue of d-orbital participation. Lewis acids also catalyze many forms of pericyclic reactions, either directly or by changing the mechanism of the reaction so that it becomes a stepwise process and, hence, no longer a true pericyclic reaction.
  • Pericyclic reactions normally show very high stereospecificity. The stereochemistry of pericyclic reactions depends on the symmetry of the interacting molecular orbitals and not on the overall symmetry of the molecules.
  • Pericyclic reactions can be frequently promoted by light as well as heat. Normally, the stereochemistry under the two sets of conditions is different, and it was thought opposite.
  • Pericyclic reactions are unusual in that very few enzymes that catalyze such reactions are known.

A set of molecular orbitals of the reactants is transformed into a corresponding set of molecular orbitals (MOs) of the products through a concerted process. If during the transformations, the symmetry of the concerted orbitals is conserved, that is, orbitals remain in phase, the reaction involves a relatively low energy transition state and is called symmetry-allowed. On the other hand, if bringing one or more orbitals out of phase destroys orbital symmetry, the transition state energy becomes very high due to an antibonding interaction, and the reaction is symmetry-forbidden (if it occurs, it will be nonconcerted).

Prior to the 1960s, organic reactivity was thought to be dominated by factors such as

  • the relative stability of reactant, transition state and product (that is, thermodynamics);
  • geometrical effects such as strain and steric interactions, hydrogen bonding,
  • electrostatic effects such as the polarity of functional groups and
  • the aromaticity of either the reactant or the product.

A new explanation was proposed based on ‘stereoelectronic’ factors, that is, on recognizing that the three-dimensional properties of the electrons and their phase relationship could dominate the other factors listed above. This theory became known as the conservation of orbital symmetry. The pericyclic reactions are highly diastereoselective under kinetically controlled conditions and are therefore of special interest in synthetic chemistry. For the same reacting systems, thermal and photochemical reactions give opposite stereochemistry.

For the sake of convenience, the pericyclic reactions are divided into five major categories: electrocyclic, cycloaddition, sigmatropic, cheletropic and group transfer reactions; of which we will discuss here three in rather greater detail than others.

11.2 ELECTROCYCLIC REACTIONS

An electrocyclic reaction is an intramolecular reaction that involves the concerted formation of a sigma bond between the two ends of a linear conjugated π (pi) system, or the reverse reaction in which the sigma bond in a cyclic reactant is broken to produce a linear conjugated π system. The product in cyclic compound has one less π bond than the reactant acyclic system.

The reactions are stereospecific. For example, cis-3,4-dimethylcyclobutene gives solely Z,E-2,4-hexadiene, which is not the thermodynamically most stable isomer.

An electrocyclic reaction is completely stereoselective. For example, when 2E,4Z,6E-octatriene undergoes an electrocyclic reaction under thermal conditions, only the cis product is formed; when 2E,4Z,6Z-octatriene undergoes an electrocyclic reaction under thermal conditions, only the trans product is formed.

However, when the reactions are carried out under photochemical conditions, the products have the opposite configuration.

Electrocyclic reactions are reversible. The cyclic compound is favoured for electrocyclic reactions that form six-member rings, whereas the open-chain compound is favoured for electrocyclic reactions that form four-member rings because of the angle strain associated with four-member rings.

Electrocyclic reactions are completely stereoselective and stereospecific. All the electrocyclic reactions are accounted for by Frontier Molecular Orbital (FMO) approach by looking only at the symmetries of the two outermost lobes of the polyene. Thus, the inner lobes may not be shown. The lobes of like sign can be either on the same side or on opposite sides of the molecule. For bond formation, the outermost lobes must rotate—a positive lobe overlapping a positive lobe or a negative lobe overlapping a negative lobe.

11.3 THEORETICAL EXPLANATION

The original explanation of Woodward and Hoffmann (1965, 1970) involved generating a so-called ‘orbital correlation diagram’ for the reaction under consideration, and then carrying out the reaction in such a manner that the symmetries of the reactant and product orbitals matched exactly. This enables a correlation diagram for the reaction to be constructed, according to the following rules: no two orbitals of the same symmetry can cross during the reaction, whilst orbitals of different symmetry can cross. The favoured pathway is the one that results in a product of the same electronic excitation as the reactant. A pathway, that results in the product in a higher electronic state than the reactant is said to be ‘forbidden’.

These correlation diagrams can be generalized for any electrocyclic reaction with appropriate symmetry. However, correlation diagrams are less readily applied for reactions with no symmetry. Dewar and Zimmerman independently noticed that the ‘topological’ properties of these correlation diagrams are very similar to those obtained using, for example, the Huckel theory for aromatic molecules. For example, the diagram for the electrocyclic conversion of hexatriene to cyclohexadiene is remarkably similar at the transition state to the ground state orbitals of benzene.

Such an approach, whilst theoretically rigorous, is not readily applicable to the majority of more complex reactions. Two much simpler methods have been outlined.

  • The Overlap of Frontier Orbitals (HOMOs and LUMOs)
  • The Concept of Transition State Aromaticity

In a practical sense, the first of these is the most easily remembered and applied. Although it may not seem obvious, this rule is actually derived from the original Woodward- Hoffmann approach. The frontier molecular method in which only the interaction between highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) is considered is the simplest and is capable of explaining the stereochemistry of almost all pericyclic reactions.

11.4 CONSERVATION OF ORBITAL SYMMETRY

In 1965 R.B. Woodward, an experimentalist, and Ronald Hoffmann, a theorist, developed the conservation of orbital symmetry theory to explain electrocyclic reactions. This theory explains the relationship among the structure and configuration of the reactant, the conditions (thermal and /or photochemical) under which the reaction takes place and the configuration of the product.

The overlap of p orbitals to form π bonds can be described mathematically using quantum mechanics. The result of the mathematical treatment can be described simply in nonmathematical terms using molecular orbital theory put forth by Kenichi Fukui.

  1. The two lobes of p orbital are of opposite phase. When two in-phase atomic orbitals overlap, a covalent bond is formed. When two out-of-phase atomic orbitals overlap, a node is created between the two nuclei. A node is a region in which there is zero probability of finding an electron.
  2. Electrons fill molecular orbitals according to the same rules that are used when they fill atomic orbitals: an electron goes into the available molecular orbital with the lowest energy, and only two electrons can occupy a molecular orbital.

Since the π-bonding portion of a molecule is perpendicular to the framework of the σ bonds, the π-bonds can be treated independently. Each carbon atom that forms a π-bond has a p orbital, and p orbitals of the carbon atoms combine to produce a π molecular orbital that is described by the linear combination of atomic orbitals (LCAO).

Ethene has only one π bond, but it has two p atomic orbitals that combine to produce two π molecular orbitals, that is, lower energy bonding π molecular orbital, designated by ψ1 and higher energy antibonding π* molecular orbital, ψ2. The bonding molecular orbital results from additive overlap of the atomic orbitals, whereas the antibonding molecular orbital results from subtractive overlap.

1,3-Butadiene has two conjugated bonds; it has four p atomic orbitals that can combine linearly in four different ways to give four π molecular orbitals: ψ1, ψ2, ψ3 and ψ4. Notice that orbitals are conserved: two atomic orbitals combine to produce two molecular orbitals; four atomic orbitals combine to produce four molecular orbitals; six atomic orbitals combine to produce six molecular orbitals.

Let us first define the symmetry properties of a 1s and a 2p orbital with respect to a plane of symmetry or an axis of symmetry. Imagine holding a mirror perpendicular to the plane of the paper along the centre of the molecular orbital. If the left half of the molecular orbital is the mirror image of the right half, the molecular orbital is symmetric. If it is not, it is antisymmetric. The easier way to know whether a molecular orbital is symmetric or antisymmetric to plane is: if the p orbitals at the ends of the molecular orbitals are identical, the molecular orbital is symmetric and if they are not identical, the molecular orbital is asymmetric.

One can take this one step further by considering the symmetry properties of molecular orbitals formed by the overlap of two or more atomic orbitals. Molecular orbitals formed from two overlapping sigma orbitals:

Molecular orbitals formed from two overlapping p orbitals (sigma bonds):

Molecular orbitals formed from two parallel overlapping p orbitals (π bonds):

We can now use these basic orbitals to construct the relevant molecular orbitals for two interconverting molecules, cyclobutene and butadiene. Note particularly that we need only construct the molecular orbitals explicitly involved in the reaction; most of the sigma framework remains unchanged and no orbitals derived from this need to be considered.

In order to interconvert cyclobutene and butadiene, the four molecular orbitals labelled ψ1, ψ2, ψ3, ψ4 must be converted into ψσ, ψπ, ψπ*, ψσ*. There are two stereochemically distinct ways in which this might be accomplished.

A molecular orbital is bonding if the number of bonding interactions is greater than the number of nodes between the nuclei, and the molecular orbital is antibonding if the number of bonding interactions is fewer than the number of nodes between the nuclei. The normal electronic configuration of a molecule is known as its ground state. If a molecule absorbs light of an appropriate wavelength an electron from its ground state HOMO (ψ2) will be promoted to its excited state LUMO (ψ3). In the excited state, the HOMO is ψ3 and LUMO is ψ4.

The product of an electrocyclic reaction is formed as a result of formation of a new σ bond. In order to form this bond, the p-orbitals at the end of the conjugated system must rotate so they can overlap head-to-head (and rehybridize to sp3). Rotation can occur in two ways. If both orbitals rotate in the same direction (both clockwise or both anti clockwise), ring closure is conrotatory. If the orbitals rotate in opposite directions, ring closure is disrotatory.

The 4π-electrons of butadiene are accommodated in the two bonding molecular orbitals, ψ1 and ψ2 so that ψ2 is the HOMO. It is quite evident that overlap of the terminal π-lobes of the same phase, forming a σ-bond is possible only by a conrotatory motion (rotation of 90° of two terminal bonds in the same direction). A disrotatory motion in the HOMO, on the other hand, leads to an antibonding interaction (σ*). In thermal electrocyclic reactions of a 4π-electron system, therefore, only conrotatory motion is allowed and the stereochemistry follows accordingly. On photoexcitation of butadiene, one of the π-electrons is transferred to ψ3 that becomes the HOMO. An inspection of the molecular orbital shows that a disrotatory motion leads to an overlap of the terminal π-lobes of the same phase and a different stereochemistry follows.

Under thermal conditions, 2E,4Z-hexadiene cyclizes to cis-3,4-dimethyl-cyclobutene, and 2E,4E-hexadiene cyclizes to trans-3,4-dimethyl-cyclobutene. When the reactions are carried out under photochemical conditions, the configuration of the product changes that is, the trans isomer is obtained instead of the cis isomer and the cis isomer is obtained instead of the trans isomer.

Some notations, which are used in cycloaddition reactions, may be defined here to indicate the mode of additions. If a component undergoes addition on the same face, it is called suprafacial component, whereas if a component undergoes addition on opposite faces, it is called an antarafacial component. The two modes of addition are known as suprafacial and antarafacial, respectively. Accordingly, conrotation involves an antarafacial and disrotation a suprafacial interaction between the two termini of the reacting species in electrocyclic reactions.

In a thermal conrotatory ring opening of cyclobutene, the σ-HOMO of the σ-component interacts with the π-LUMO of the π-component with in-phase overlap, the former behaving as a suprafacial component and the latter as an antarafacial one. The reaction is designated [σ2s + π2a]. Alternatively, the same reaction may be considered in terms of interaction between the σ-LUMO, that is, σ* and the π-HOMO, that is, π and thus designated [σ2a + π2s]. Similarly, the disrotatory photochemical ring opening of cyclobutene, in which the σ-HOMO interacts suprafacially with the π-LUMO*, is designated [σ2s + π2s]. Alternatively, if the interaction between the σ-LUMO and the π-HOMO is considered, it may be designated [σ2a + σ2a].

A symmetry-allowed pathway leads to overlap of in-phase orbitals; a symmetry-forbidden pathway leads to overlap of out-of-phase orbitals. Symmetry-allowed reaction could take place under relatively mild conditions. If a reaction is symmetry-forbidden, it cannot take place by concerted pathway.

Opening of the cyclopropyl cation involves only one electron arrow and, hence, the number of cyclically conjugated electrons is 2 (4n+2, n=0). Under thermal conditions, this will go via a Huckel topology involving suprafacial components, which has the effect of rotating both methyl groups outwards in opposite directions (disrotation).

Whilst this rule is normally followed fairly well for ground states, it can be overturned when, for example, steric or geometrical strain in the ‘allowed’ pathway promotes the ‘forbidden’ route. The situation is actually more complex for photochemical reactions, and much recent evidence suggests that the Woodward-Hoffmann rules are not always followed.

Thus a ‘suprafacial’ or ‘Huckel’ transition state in a pericyclic reaction is particularly favourable if the number of cyclically conjugated π electrons in the transition state equals 4n+2 (the Huckel rule, where n = 0, 1, 2, etc).

The treatment is extended to 1,3,5-hexatriene–cyclohexadiene interconversion. The ground state HOMO (ψ3) of a compound with three conjugated π-bonds, such as (2E,4Z,6E)-octatriene, is symmetric. This means that ring closure under thermal conditions is disrotatory. In disrotatory ring closure of (2E,4Z,6E)-octatriene, the methyl groups are both pushed up (or down) resulting in the formation of a cis product.

In disrotatory ring closure of (2E,4Z,6Z)-octatriene, the one methyl group is pushed up and the other is pushed down resulting in the formation of a trans product.

If the reaction takes place under photochemical conditions, we must consider the excited state HOMO rather than the ground state HOMO. The excited state HOMO (ψ4) of a compound with three π-bonds is asymmetric. Therefore, under photochemical conditions, (2E,4Z,6Z)-octatriene undergoes conrotatory ring closure, so both methyl groups are pushed down (or up) and the cis product is formed.

Thus, symmetry of the HOMO of the compound undergoing ring closure controls the outcome of an electrocyclic reaction. Ring closure of (2E,4Z)-hexadiene forms cis-3,4-dimethylcyclobutene. The compound undergoing ring closure has two conjugated π-bonds. The ground state HOMO of a compound with two conjugated π-bonds is asymmetric, so ring closure is conrotatory. Conrotatory ring closure of (2E,4Z)-hexadiene leads to the cis product.

However, if the reaction is carried out under photochemical conditions, the excited state HOMO of a compound with two conjugated π-bonds is symmetric. So 2E,4Z-hexadiene will undergo disrotatory ring closure, resulting in the trans product, while 2E,4E-hexadiene will undergo disrotatory ring closure and form the cis product.

We have seen that the HOMO of a compound with two conjugated double bonds is asymmetric, whereas the HOMO of a compound with three conjugated double bonds is symmetric. If we examine molecular orbital diagrams for compounds with four, five, six and more conjugated double bonds, we can conclude that the ground state HOMO of a compound with an even number of conjugated double bonds is asymmetric, whereas the ground-state HOMO of a compound with an odd number of conjugated double bonds is symmetric. Therefore, from the number of conjugated double bonds in a compound, we can immediately tell whether ring closure will be conrotatory (an even number of conjugated double bonds) or disrotatory (an odd number of conjugated double bonds) under thermal conditions. However, if the reaction takes place under photochemical conditions, everything is reversed, because the ground-state and excited-state HOMOs have opposite symmetries.

We have seen that the stereochemistry of an electrocyclic reaction depends on the mode of ring closure. In turn, the mode of ring closure depends on the number of conjugated π bonds in the reactant and on whether the reaction is carried out under thermal or photochemical conditions. What we have learned about pericyclic reactions can be summarized with a series of selection rules. These are also known as the Woodward–Hoffmann rules for electrocyclic reactions.

 

Table 11.1 Woodward–Hoffmann Rules for Electrocyclic Reactions

No. of conjugated π bond Reaction conditions Allowed mode of ring closure
Even number Thermal
Photochemical
Conrotatory
Disrotatory
Odd number Thermal
Photochemical
Disrotatory
Conrotatory

The following series of reactions illustrate just how easy it is to determine the mode of ring closure and, therefore, the product of an electrocyclic reaction. The reactant of the first reaction has three conjugate double bonds and is undergoing ring closure under thermal conditions. Ring closure is therefore disrotatory. Disrotatory ring closure of this reactant causes the hydrogens to be cis in the ring-closed product.

The second step is a ring-opening electrocyclic reaction that takes place under photochemical conditions. Because of the principle of microscopic reversibility, the orbital symmetry rules used for a ring-closure reaction also apply to the reverse ring-opening reaction. Because the reaction occurs under photochemical conditions, ring opening (or ring closure) is conrotatory. In order for conrotatory rotation to result in a product with cis hydrogens, the hydrogens in the compound undergoing ring closure must point in the same direction. The third step is a thermal ring closure of a compound with three conjugated double bonds, so ring closure is disrotatory. Drawing the hydrogens and the arrows allows us to determine the configurations of the hydrogens in the ring-closed product.

Huckel was also able to show that if a cyclic conjugated π system is irradiated with light so that it goes into the first ‘excited’ electronic state, it is especially stable if the number of cyclically conjugated electrons equal 4n. Hence, photochemically activated pericyclic reactions will proceed suprafacially via a Huckel transition state if the electron count corresponds to 4n. The antarafacial mode described above has no counterpart in a stable aromatic molecule, as benzene did for the suprafacial mode. An antarafacial mode could be formed by taking benzene and giving the π system a 180° twist. If this is done, the resultant ‘twisted’ benzene has been termed ‘Mobius Benzene.’

Here the electron density continuity passes through the plane of the molecule, from the top face through to the bottom face. Such a MOBIUS system it turns out is especially stable if it contains 4n electrons (or 4n+2 if photochemically excited). These four conditions are most easily summarized as follows:

  • A thermally activated pericyclic reaction will proceed via a Huckel topology containing only suprafacial components if the cyclically conjugated π electrons equal 4n+2 (n=0,1,2 etc).
  • A photochemically activated pericyclic reaction may proceed via a Huckel topology containing only suprafacial components if the cyclically conjugated π electrons equal 4n.
  • A thermally activated pericyclic reaction will proceed via a Mobius topology containing ONE antarafacial component if the cyclically conjugated π electrons equal 4n (n = 0,1,2 etc).
  • A photochemically activated pericyclic reaction may proceed via a Mobius topology containing ONE antarafacial component if the cyclically conjugated π electrons equal 4n+2 (n = 0,1,2 etc).

It would be a good point here to remind you that, although all electrocyclic reactions are allowed both thermally and photochemically, providing the reaction is right, the steric requirements for conrotatory or disrotatory cyclization or ring opening may make one or both modes impossible.

11.5 CYCLOADDITION REACTIONS

In a cycloaddition reaction, two different π-bond-containing molecules react to form a cyclic molecule. The reverse process is known as cycloreversion or retro-cycloaddition. The Diels-Alder reaction is one of the best known examples of a cycloaddition reaction. Cycloaddition reactions are classified according to the number of π electrons that interact in the reaction. The Diels-Alder reaction is a [4+2] cycloaddition because one reactant has four interacting π electrons and the other reactant has two interacting π electrons. Only the π electrons participating in electron rearrangement are counted. If the number of π electrons in the two components is m and n respectively, the cycloaddition is called [m+n] addition. Cycloaddition may be intermolecular- or intramolecular depending on whether the two components belong to different or the same molecule.

11.6 FRONTIER MOLECULAR ORBITAL APPROACH

In a cycloaddition reaction, the orbitals of one molecule must overlap with the orbitals of the second molecule. Therefore, the frontier molecular orbitals of both reactants must be evaluated to determine the outcome of the reaction. Since the new σ bonds in the product are formed by donation of electron density from one reactant to the other, we must consider the HOMO of one of the molecules and the LUMO of the other, because only an empty orbital can accept electrons. It does not matter which reacting molecule’s HOMO is used. It is required only that we use the HOMO of the one and the LUMO of the other.

There are two modes of orbital overlap for the simultaneous formation of two σ bonds: suprafacial and antarafacial. Bond formation is suprafacial if both σ bonds form from the same side of the π system. Bond formation is antarafacial if the two σ bonds form from opposite sides of the π system. Suprafacial bond formation is similar to syn addition whereas antarafacial bond formation resembles anti addition.

A cycloaddition reaction that forms a four-, five-, or six-member ring must involve suprafacial bond formation. The geometric constraints of these small rings make the antarafacial approach highly unlikely, even if it is symmetry-allowed (the overlapping orbitals are in phase). Antarafacial bond formation is more likely in cycloaddition reactions that form larger rings.

Frontier orbital analysis of a [4+2] cycloaddition reaction shows that overlap of in-phase orbitals to form the two new σ bonds requires suprafacial orbital overlap. This is true whether we use the HOMO of the alkene (a system with one π bond) and the LUMO of the diene (a system with two conjugate π bonds) or the LUMO of the alkene and the HOMO of the diene. Now we can understand why Diels-Alder reactions occur with relative ease. It is clear that in both combinations, there are bonding interactions at the termini. The [π2s+π4s] addition is thus thermally or symmetry allowed. Since suprafacial–suprafacial and antarafacial–antarafacial combinations give the same prediction, [π4a + π2a] addition is also thermally allowed. In photochemical reaction, the reverse is true, that is, [π2a + π4s] or [π4a + π2s] addition takes place.

A [2+2] cycloaddition reaction does not occur under thermal conditions but does take place under photochemical conditions.

The reason for this is apparent from an examination of the frontier molecular orbitals. Under thermal conditions, suprafacial overlap is not symmetry-allowed (the overlapping orbitals are out of phase). Antarafacial overlap is symmetry-allowed, but is not possible because of the small size of the ring. Under photochemical conditions, the reaction can take place because the excited-state HOMO has symmetry opposite that of the ground state HOMO. Therefore, overlap of the excited-state HOMO of one alkene with the LUMO of the second alkene involves symmetry-allowed suprafacial bond formation.

Notice that in the photochemical reaction, only one of the reactants is in an excited state. Owing to the very short lifetimes of excited states, there is little likelihood that two excited states will find one another to interact.

Similar treatment of the cycloaddition of two butadiene molecules - an 8π electron (m + n = 8) system shows that [π4s + π4a] addition is thermally allowed and [π4s + π4s] is photochemically allowed. A [6+2] addition behaves in the same fashion. A [6 + 4] cycloaddition behaves exactly in the same way as the [4 + 2] addition.

Woodward–Hoffmann rules In a thermal pericyclic reaction the total number of (4q + 2)s and (4r)a components must be odd. A component is a bond or orbital taking part in a pericyclic reaction as a single unit. A double bond is a π2 component. The number 2 is the most important part of this designation and simply refers to the number of electrons. The prefix π tells us the type of electrons. A component may have any number of electrons (a diene is a π4 component) but may not have mixtures of π and σ electrons. Now look back at the rule. Those mysterious designations, (4q+2) and (4r), simply refer to the number of electrons in the component where q and r are integers. An alkene is a π2 component and so it is of the (4q+2) kind, a diene is a π4 so is of the (4r) kind.

A few relevant points emerge from the above discussion:

  1. For a two-component cycloaddition, the maximum number of modes of addition is 22 (for n components, it is 2n): (s,s), (a,s), (s,a) and (a,a).
  2. Only in the (s,s) mode of addition do the two π systems approach in parallel plane. In all other modes of addition, the components approach orthogonally.
  3. Configuration of groups at the two termini of a suprafacial component is retained, but that on a antarafacial component is inverted.
  4. For either m or n greater than 2, there are two modes of (s,s) additions giving endo- and exo-products.

Now what about the suffixes ‘s’ and ‘a’? The suffix ‘s’ stands for suprafacial and ‘a’ for antarafacial. A suprafacial component forms new bonds on the same face at both ends, whereas an antarafacial component forms new bonds on opposite faces at both ends. See how this works for the Diels-Alder reaction. Here is the routine.

  1. Draw the mechanism for the reaction (we shall choose a general one)
  2. Choose the components. All the bonds taking part in the mechanism must be included and no others
  3. Make a three-dimensional drawing of the way the components come together for the reaction, putting in orbitals at the ends of the components (only!)
  4. Join up the components where new bonds are to be formed. Coloured dotted lines are often used.
  5. Label each component s or a depending on whether new bonds are formed on the same or on opposite sides.
  6. Count the number of (4q+2)s and (4r)a components. If the total count is odd, the reaction is allowed. There is one (4q+2)s component (the alkene) and no (4r)a components, total = 1, so it is an allowed reaction.
  • The cycloadditions that do occur thermally, for example, the Diels-Alder reaction, have (4n+2π) electrons in their ‘aromatic’ transition states.
  • The cycloadditions that do not occur thermally, for example, the dimerization of alkenes and of dienes, has 4nπ electrons in their ‘anti-aromatic’ transition states.
11.7 SIGMATROPIC REARRANGEMENTS

The last class of concerted pericyclic reactions that we will consider is the groups of reactions known as sigmatropic rearrangements. In a sigmatropic rearrangement, a σ bond in a molecule is broken, a new σ bond is formed and the π electrons rearrange. The σ bond that breaks is bonded to an allylic carbon. It can be a σ bond between a carbon and hydrogen, between a carbon and another carbon, or between a carbon and an oxygen, nitrogen, or sulphur. ‘Sigmatropic’ comes from the Greek work tropos, which means, ‘change,’ so sigmatropic means ‘sigma-change’. The following points may be noted:

  • The number of π bonds does not change; both the reactant and the product contain the same number of π bonds.
  • The σ bond that cleaves can be in the middle of the π system or at the end of the π system.
  • The σ bond that breaks is bonded to an allylic carbon.
  • The numbering system used to describe a sigmatropic rearrangement [i, j] differs from any numbering system you have seen previously. First, mentally break the σ bond in the reactant into two pieces, and then assign number 1 to both the atoms involved in this bond. Then, the atoms in each direction from the bond being broken is counted in each of the fragments that connect the broken σ bond and the new σ bond. The two numbers are put in brackets with the small number stated first. For example, in the [2,3] sigmatropic rearrangement shown below, two atoms (N, N) connect the old and new σ bonds in one fragment and three atoms (C, C, C) connect the old and new σ bonds in the other fragment.

There are two different types of sigmatropic reactions, a) those that involve the migration of a hydrogen atom and b) those that involve a carbon or other atom. In the former category, the hydrogen atom can migrate either suprafacially or antarafacially across the conjugated system, leading to Huckel or Mobius topology for the transition states.

Something different can occur when a carbon migrates.

The configuration at the migrating centre can be either retained (= suprafacial mode = Huckel topology) or inverted (= antarafacial mode = Mobius topology). Note that Mobius transition states are relatively common in this class because there is often little strain involved in inversion of configuration at a carbon. The most common category of hydrogen shift involves a so called [1,5] sigmatropic shift.

In the transition state of a sigmatropic rearrangement, the group that migrates is partially bonded to the migration source and partially bonded to the migration terminus. There are two possible modes for rearrangement. If the migrating group remains on the same face of the π system, the rearrangement is suprafacial. If the migrating group moves to the opposite face of the π system, the rearrangement is antarafacial.

Sigmatropic rearrangements have cyclic transition states. If the transition state has six or fewer atoms in the ring, rearrangement must be suprafacial because of the geometric constraints of small rings.

 

Table 11.2 Woodward–Hoffmann Rules for Sigmatropic Rearrangements

No. of pairs of electron in the reacting system Reaction conditions Allowed mode of ring closure
Even Number Thermal
Photochemical
Antarafaciala
Suprafacial
Odd number Thermal
Photochemical
Suprafacial
Antarafaciala
a Although antrafacial ring closure is symmetry-allowed, it can occur only with large rings.

 

A [1,3] sigmatropic rearrangement involves a π bond and a pair of σ electrons, or we can say that it involves two pairs of electrons. A [1,5] sigmatropic rearrangement involves two π bonds and a pair of σ electrons (three pairs of electrons), and a [1,7] sigmatropic rearrangement involves four pairs of electrons. We can use the same symmetry rules for sigmatropic rearrangements that we used for cycloaddition reactions if we substitute ‘number of pairs of electrons’ for ‘sum of the number of π bonds.’ Recall that the ground-state HOMO of a compound with an even number of conjugated double bonds is asymmetric, but the ground-state HOMO of a compound with an odd number of conjugated double bonds is symmetric.

A Cope rearrangement is a [3,3] sigmatropic rearrangement of a 1,5-diene. A Claisen rearrangement is a [3,3] sigmatropic rearrangement of allyl vinyl ether. Both rearrangements form six-member ring transition states. Therefore, the reactions must be able to take place by a suprafacial pathway. Whether or not a suprafacial pathway is symmetry-allowed depends on the number of pairs of electrons involved in the rearrangement. Because [3,3] sigmatropic rearrangements involve three pairs of electrons, they occur by a suprafacial pathway under thermal conditions. Therefore, both Cope and Claisen rearrangements readily take place under thermal conditions.

11.8 SUMMARY

The pericyclic reactions considered in this chapter are of three general types: electrocyclic, cycloaddition and sigmatropic reactions. A cycloaddition requires the interaction of the π system from two molecules, whereas an electrocyclic reaction takes place by the interaction of the π orbitals at the ends of a single π system within one molecule. A sigmatropic shift describes the migration of a σ bond across a π system. Two classes of sigmatropic shifts that achieve specific skeletal rearrangements are discussed: in the Claisen and the Cope rearrangements, both ends of a σ bond shift by three carbons, forming a new σ bond at those positions and producing a rearranged backbone. A heteroatom can be incorporated at the σ bond, at a multiple bond, or as a substituent on the carbon framework.

PROBLEMS
  1. (a) Suggest a mechanism for the following reaction that explains the observed stereochemistry:

    (b) Propose a structure for 1 consistent with the spectral evidence and classify the type of pericyclic reaction occuring, paying particular attention to the expected stereochemistry of the product. 1H NMR includes the following: δ 1.43 (3H, s), 1.52 (3H, triplet, J=1.5Hz), 3.76 (1H, multiplet), 5.71 (1H, multiplet), 5.73 (1H, double doublet, J=10, 1.5Hz), 5.79 (1H, doublet, J=10Hz), 6.74 (1H) + phenyl protons. No carbonyl peak is evident from the infrared spectrum and no cyclopropyl group is present in 1.

  2. When compound 1 is heated, a transient intermediate 2 is formed, rapidly rearranging to two new compounds 3 and 4. An aqueous workup of this mixture, in which traces of acid are present, shows that 4 has been converted to 5. Suggest structures for compound 2-4, and mechanisms for all the steps in this sequence.
  3. Photolysis of compound 7 in benzene as solvent produces the triene 8, which then decomposes thermally to give butadiene and tetralin (12), via, it is thought, the intermediates 9, 10 and 11. Propose and classify the mechanisms for these transformations, and suggest a structure for 11.
  4. (a) When molecules 1 and 2 (E = CO2Me) are heated together, the transient intermediates 3 and 4 are formed, with the eventual isolation of 5 and 6 as the two final products. Propose structures for 3 and 4 and a mechanism for the formation of 5 and 6.

    (b) Propose structures for A and B and mechanisms for the following transformations.

  5. (a) Propose a mechanism for the following transformation, identifying any intermediates that might be involved:

    (b) The following transformation can take place via two alternative pathways, one involving intermediates 2 and 3, the other involving a single intermediate 4. When 13C label is introduced into the reactant, the product can be shown to have the label distributed as shown. Suggest possible structures for the intermediates 2-4, and on the basis of the labelling information, indicate which mechanistic pathway is actually followed.

  6. Classify the type of thermal reaction occuring in the sequences below, indicating clearly any stereochemical implications.
  7. Propose a mechanism for the following transformation, including an explanation for the formation of two products and indicating the number of electrons involved in each step. Extrapolating from this explanation, suggest a related third product that might have formed, but, as it happens, was not detected with these particular substituents.
  8. Propose a mechanism for the following transformation:
  9. How would you employ pericyclic reactions in the synthesis of the following?
  10. Rationalize the following thermal reactions:
  11. Draw the structure of the chief product of each of the following thermal reactions:
  12. Classify each of the following transformations as a cycloaddition, an electrocyclic reaction, a sigmatropic rearrangement, or a nonpericyclic reaction.
OBJECTIVE TYPE QUESTIONS
  1. Ionic reactions generally
    1. occur in the gas phase
    2. occur in nonpolar solvents
    3. occur at a high temperature
    4. occur in solution in polar solvents
  2. Diels-Alder reaction is a
    1. cycloaddition reaction
    2. electrocyclic reaction
    3. sigmatropic reaction
  3. Thermal electrocyclic reactions involve
    1. (4n + 2) π electrons and are disrotatory
    2. (4n + 2) π electrons and are conrotatory
    3. (4n) π electrons and are disrotatory
    4. (4n) π electrons and are conrotatory
  4. If a nucleophile reacts with an electrophile the reaction is said to be
    1. a polar reaction
    2. a radical reaction
    3. a pericyclic reaction
    4. none of the above
  5. Pericyclic reactions are
    1. highly stereoselective
    2. highly stereospecific
    3. highly regioselective
    4. highly regiospecific
  6. Cope rearrangement is
    1. an intramolecular process proceeding through a cyclic transition state
    2. an intermolecular process proceeding through a cyclic transition state
    3. an intramolecular process proceeding through a non-cyclic transition state
    4. an intermolecular process proceeding through a non-cyclic transition state
  7. In a sigmatropic rearrangement
    1. a σ bond is broken in the reactant, a new σ bond is formed in the product and the π bonds rearrange
    2. a σ bond is broken in the reactant and a new π bond is formed in the product
    3. a π bond is broken in the reactant, a new σ bond is formed in the product and the π bonds rearrange
    4. a π bond is broken in the reactant and a new σ bond is formed in the product
  8. A molecular orbital is termed as bonding
    1. if the number of bonding interactions is greater than the number of nodes between the nuclei
    2. if the number of bonding interactions is fewer than the number of nodes between the nuclei
    3. if the number of bonding interactions is equal to the number of nodes between the nuclei
  9. Frontier orbitals are
    1. HOMO and LUMO
    2. HOMO only
    3. LUMO only
    4. none of the above
  10. In the excited state of 1,3,5-hexatriene the HOMO is
    1. ψ3
    2. ψ2
    3. ψ4
    4. ψ5