Advanced organic chemistry wim dehaen

WIM DEHAEN

ADVANCED ORGANIC CHEMISTRY


Chapter 1 Concerted reactions

During concerted reactions the cleavage of the bonds of the starting materials and the
formation of the new bonds of the product happen at the same time (in other words in concert)
without the occurrence of discrete intermediates. A very important class of concerted
reactions is formed by the pericyclic reactions. The latter are characterized by a cyclic
transition state. In the text below we will discuss the different types of pericyclic reactions at
length. In a second part of the chapter others examples of concerted reactions are given,
together with the consequences for the stereochemistry of the products formed.

1.1 Pericyclic reactions : properties and types

-During the course of the reaction no (high-energy) radical, carbocation or carbanion
intermediates are formed. In many cases, the activation energy will be rather low as a
consequence. In general, there are no important solvent effects observed in these reactions
because during the reaction no (large) changes in polarity occur.
-The cyclic transition state implies a large degree of organisation of the reagents, so the
reaction entropy will be negative.
-The pericyclic reactions will in many cases lead to the stereo- and regioselective formation of
products even if several isomers would be possible.
-The reactions are activated by heating (thermally) or by irradiation with UV- or visible light
(photochemically).

R

h

+
R

photochemical [2+2]cycloaddition

R

Synthesis of cyclopropanes from carbenes

+

+
S
O2

thermal Diels-Alder cycloaddition

R

SO2

transformation of sulfolene to butadiene and SO2

1


We can distinguish three types of pericyclic reactions:

-Cycloadditions: two separate molecules or fragments form a new cyclic system, and during

this process two -bonds disappear and two -bonds are formed. An example is the
photochemical [2+2] dimerisation of alkenes to form cyclobutanes or the thermal [4+2] DielsAlder cycloaddition reaction. Cheletropic reactions and the reverse process, the extrusion
reactions, form a special case in which the two -bonds are formed (respectively cleaved) at
the same atom. These [n+1] processes will for instance take place for the addition of carbenes
(see later) to alkenes and the formation of butadiene and SO2 from sulfolene.
-Electrocyclic reactions: within a single, conjugated open chain system with n -bonds a
transformation occurs to a cyclic system with (n-1)  bonds and one (1) newly formed bond. In function of the reaction circumstances, the reverse reaction (ring opening) may take
place. The reaction takes place thermally or photochemically.

cyclobutene

butadiene

-Sigmatropic rearrangements: during the reaction, a group R migrates over a conjugated system, of which the bonds shift during the migration. Thus, the total amount of - or π-bonds
does not change during these reactions. An example is the Claisen rearrangement, in which an
allyl group shifts over an enolate system, resulting in the formation of an unsaturated carbonyl
compound. This is an example of a [3,3]-sigmatropic rearrangement.

O
O

Claisen rearrangement

2


1.2 Pericyclic reactions : general principles

1.2.1 Molecular orbitals

Molecular orbitals are obtained by linear combination of atomic orbitals (LCAO). Atomic
orbitals can be seen as wave functions, combining in-phase (bonding interaction) or out-ofphase (antibonding interaction). If two p-orbitals are combined following the long axis, this
results in the formation of a bonding -orbital and an antibonding *-orbital. The latter has a
higher energy and the orbitals with the lowest energy are the first to be filled with electrons.
These two simple orbitals are symmetric in relation to the bond axis, while in regard to the
nodal plane (m, the plane perpendicular to the bond axis) the -orbital is symmetric (S) and
the *-orbital antisymmetric (A). In relation to the C2-axis perpendicular to the bond axis this
is the same: the -orbital is symmetric (S) and the *-orbital antisymmetric (A).
The - and *-orbitals are formed by lateral overlapping (respectively bonding and
antibonding) of two p-orbitals. These orbitals are both antisymmetric in regard to the bond
axis, and in relation to the nodal plane m the -orbital is symmetric and the *-orbital
antisymmetric. In relation to C2 this situation is reversed.



Energy













C=C bond

lateral overlap

C-C bond
axial overlap

The wave function 1 = c11 + c22 for the bonding - and -orbitals,
and the wave function 2 = c11 - c22 for the antibonding *- and *-orbitals.

3


The numbers c1 and c2 are the orbital coefficients. Visually, these coefficients are shown by
the size of the orbital lobes. For symmetric compounds (e.g. ethene) c1 = c2, in other cases
(e.g. CH2=O) the two coefficients are different.
Ethene has both (*)- and (*)-orbitals. The energy of the - en *-orbitals is given in
theoretical discussions as respectively + and -, in which  is the energy of the original
p-orbital and  the energy difference by delocalisation of the electrons over the two atoms of
the molecule. Both  and  are negative energy values.
The -orbital is in this case the highest occupied molecular orbital (HOMO), and the *orbital is the lowest unoccupied molecular orbital (LUMO). Both are the frontier orbitals.



Energy







LUMO



HOMO


Electronic configuration of ethene

In linearly conjugated systems there are several (>2) p-orbitals that simultaneously enter in
lateral interaction with each other. The electrons of the resulting molecular orbitals are
delocalised over all the participating atoms. A prerequisite is that the conjugated system is
not interrupted by sp3-hybridised atoms. Atomic orbitals that are perpendicular (as in allenes
or cumulenes) can not overlap and are not conjugated. Examples of simple linearly
conjugated systems are butadiene (n = 4) and allyl (n =3) (cation, radical or anion). 1,4Pentadiene has two localised double bonds, therefore it is not conjugated.

4


isolated double bonds

conjugated systems

butadiene
1,4-pentadiene

CH2

allyl anion

H2C

C

CH2

allene

The n different wave functions of a system with n atoms are described according to LCAO for
the j-th orbital as:
j = c1j1 + c2j2 + c3j3 +... + cnj3

The coefficients for polyene systems with n atoms can be theoretically calculated (the socalled Hückel approach) whereby a coefficient crj of the r-th atom orbital in the j-th molecular
orbital is given by:
crj = (2/n+1)0.5 x sin rj/n+1

Example: the coefficient for the third atomic orbital in the fourth wave function of a four atom
system is 0.6.
and the energy of a molecular orbital j is given in general by
E =  + m in which m = 2 cos(j/n+1). If m = 0 the orbital is non-bonding.

This approach gives information on the relative contribution of the atomic orbitals in a certain
molecular orbital (size of lobes = orbital coefficients) and also shows if the interaction is
bonding, antibonding or not-bonding. At the same time the amount of knots (electron density
= 0), and their position in the molecule, can be determined.
Application of these formulas on ethene (n =2) leads to m = 1 and c1 = c2 = 0.707.
The following system is this with n = 3, the allyl system. In this case we have three molecular
orbitals 1, (E =  + 1.414), 2 (E = ) and 3 (E =  - 1.414). Thus, the molecular orbital
2 is non-bonding.
5

An allyl cation has electron configuration 1220, an allyl radical 1221, and an allyl anion
1222. The allyl group is bent because the central carbon atom has sp2-hybridisation and thus
the angle is 120°.

The orbital coefficient c22 = 0, in other words a knot is localised on the central atom of the
second orbital of the allyl system. The other two coefficients are c21 = c12= c32 = c23 = 0.707
and c11 = c31 = c13 = c33 = 0.5. The molecular orbital 2 is the LUMO for the allyl cation, and
the HOMO for the allyl anion. The molecular orbital 1 has no knots, and the molecular
orbital 3 has two knots, in between atoms 1-2 and 2-3. In general, a linearly conjugated
system in the n-th molecular orbital has n-1 knots.

Symmetry
m

Energy







C2


S

A

A

S

S

A





Molecular orbitals of allyl

The most stable conformation of butadiene (n = 4) is a zigzag structure. With LCAO four
molecular orbitals can be formed, in which four -electrons are accommodated. Thus, the
HOMO is the 2-orbital (one knot) and the LUMO is the 3-orbital (two knots). The
difference in energy between HOMO and LUMO is for butadiene (n = 4) 1.236, this is less
than the “HOMO-LUMO-gap” for the allyl cation (n = 3, 1.414) or ethene (n = 2, 2). Thus,
the longer is the conjugation, the smaller is the distance between HOMO and LUMO.
The Hückel calculations predict two orbital coefficients 0.6 and 0.371. In the two frontier
orbitals the coefficients on the two outer atoms is larger than those on the central. In the
different molecular orbitals of butadiene the knots are always located between the carbon

6


atoms, and this is typical for linearly conjugated systems with an even amount of carbon
atoms.

Furthermore, the two occupied molecular orbitals 1 and 2 show respectively a bonding and
antibonding interaction between the central atomic orbitals on C-2 and C-3. The relevant
coefficients are larger for 1 which makes the interaction more bonding. Thus, we can say
that the C-2-C-3 bond in butadiene has partial double-bond-character.
We would like to mention that in simplified representations of the molecular orbitals of
conjugated systems often all orbitals are shown with the same coefficients. It is important to
keep in mind that this does not completely correspond to reality.

Symmetry

Energy








m

C2



A

S



S

A



A

S

S

A



Below are shown the simplified representations (orbitals, energies, symmetry) of the next
homologous systems with n = 5 (pentadienyl) and n = 6 (1,3,5-hexatriene), following the
same logic. The HOMO-LUMO gap is further reduced, respectively to  (n = 5 ) and 0.890
(n = 6).

7


Symmetry





C2

S

A

A

S

S

A

A

S

S

A




Energy

m













Symmetry

Energy





m

C2

A

S





S

A

A

S



S

A



A

S

S

A










8


For cyclic conjugated systems other rules apply. The Hückel orbital theory describes the
energy of planar polycyclic polymethines (CH)n ([n]annulenes) as:
E =  + 2 cos 2r/n

with n = number of C-atoms ; r = 0, 1, 2, ...n-1

Mnemotechnically, one can obtain the energy levels by representing the molecule as a regular
polygon that is circumscribed by a circle with diameter 4. The lowest atom (situation for r =
0) should always be placed at the bottom of the circle, and the corresponding lowest energy
level is  + 2. A difference with the linear polymethines is that molecular orbitals with the
same energy (degenerate systems) can occur. In the figure below, the Hückel energy levels
are given for planar, cyclic conjugated systems of n = 3 to n = 8.
































Carrying out the calculation for a six-membered ring (benzene) shows the occurrence of 6
orbitals with r = 0, 1 , 2 , 3, 4, 5. The Hückel energies are respectively  +2, +, -, -2,
- and +.

9


1.2.2 Aromaticity

Hückel’s rule : Planar, fully conjugated systems with (4n + 2)  electrons have all binding
orbitals filled and thus are very stable. These systems are aromatic. The analogous systems
with 4n  electrons are anti-aromatic (n is in both cases an integer).
Aromatic systems are significantly more stable in comparison with the linear analogs and
have a diamagnetic ring current. Anti-aromatic systems are less stable than the linear analogs
and the system will assume a non-aromatic structure whenever possible, for instance by loss
of planarity as in cyclooctatetraene.
This rule can be further explained after a closer look at the energy levels in the figures above
and after a comparison of the stabilisation energies of the filled orbitals of the cyclic and noncyclic systems.
We can define for cyclic polyenes a Hückel system in which the base orbital, in other words
the lowest filled -level (1) has p-lobes that overlap in-phase.
On the other hand, in a Möbius- or anti-Hückel-system one end of the chain has been turned
over 180° (or n), so we have a phase dislocation. These definitions can be expanded by
stating that a system with an even amount of phase dislocations is a Hückel system, and a
system with an odd amount of phase dislocations is a Möbius system.

Hückel-system

Möbius-system

A so-called Möbius ring can be prepared by turning a strip of paper at one end over 180° and
then joining the ends. Note that a Möbius ring has only one side.

10


Evidently, such twisted compounds have large strain, making them unstable. Therefore,
Möbius systems have never been isolated, but are rather of theoretical interest to describe the
transition states of pericyclic reactions.
Hückel systems as before are aromatic with 4n +2 -electrons, Möbius systems on the other

hand are aromatic when they possess 4n -electrons.

1.2.3 Aromaticity principle for the description of pericyclic reactions

This approach was first used by Zimmerman and Dewar on cyclic transition states in
pericyclic reactions. These transition states can be seen as aromatic (favourable) or antiaromatic (unfavourable). The derivated rule is the following :

Pericyclic reactions occur thermally (are allowed) when an aromatic transition state can be
formed.
This aromatic transition state is attained for a Hückel system with 4n +2 -electrons or a
Möbius system with 4n -electrons. For photochemical processes that occur via the lowest
excited state, this rule is reversed : the allowed processes are Hückel systems with 4n electrons or Möbius systems with 4n + 2 -electrons.
A few pointers when applying this aromaticity rule:
-In the transition state the base orbitals are used (ground orbitals of the reacting systems, -,
p- of -orbitals) with the corresponding phase signs. (Do not use frontier orbitals !)
-the number of electrons and the number of phase dislocations are determined.
-from these data can be determined if the reaction is allowed or not.

1.2.4. Frontier orbital approach

During chemical reactions, and especially pericyclic reactions, the process of overlapping
between the filled orbitals of a substrate and the empty orbitals of a reagent (and vice versa)
determines the course of the reactions.

11


The result of an interaction between two filled orbitals is repulsive because the combination
leads to a bonding and antibonding orbital that are both occupied. The resulting energy effect
is unfavorable. The destabilisation by the antibonding orbital is larges than the stabilisation

caused by the bonding orbital because of the coulombic repulsion of the two atoms. Empty
orbitals of two reagents have no stabilising effect because they contain no electrons.

LUMO-1

LUMO-1

LUMO-2

LUMO-2

HOMO-1

HOMO-1

HOMO-2

HOMO-2

LUMO-1
LUMO-2

HOMO-1
HOMO-2

The interaction between filled and empty orbitals will be stronger (leads to more stabilisation,
lowering of energy) if these orbitals are closer to each other in energy. Therefore, it is mainly
the frontier orbitals (HOMO and LUMO) that will have an influence on the chemical reaction.
Electron poor reagents have a relatively low LUMO and will specifically use this frontier
orbital in their reactions. Electron rich products have a relatively high HOMO, giving the

strongest interactions.

12


The frontier orbital approach states that HOMO and LUMO, other than being close in energy,
should also have a comparable symmetry. The symmetry of the two frontier orbitals should be
such that the two ends combine in a bonding interaction (the same phase sign).

LUMO

HOMO

bonding interaction

LUMO

HOMO

antibonding interaction

1.2.5 Woodward-Hoffmann rules
Fukui and Hoffmann obtained the Nobel prize in 1981 for their theoretical application of
orbital symmetry to pericyclic reactions. Woodward, who co-developed this approach, had
already died in 1979 but did obtain the Nobel prize in 1965 for his synthetic work. A
summary of this work is given by the Woodward-Hoffmann rules:

In a thermal pericyclic reaction the total amount of (4q+2)s and (4r)a components should be
odd.

This short sentence needs some further explanation. The components mentioned are bonds or
orbitals that participate in a pericyclic reaction as a separate unit. The 4q+2 and 4r refer to the
number of participating electrons, q are r integer, in most cases 0, 1 or (sometimes) 2. The
suffixes “s” and “a” refer to a suprafacial, respectively a antarafacial component. For a

13


suprafacial component, the new bonds are formed on the same side of the component, and for
an antarafacial component the new bonds are formed on opposite sides.

1.3 Cycloadditions

1.3.1 Diels-Alder reaction

The most famous cycloaddition reaction is the Diels-Alder reaction. This is a concerted [4+2]
cycloaddition, in which 4 and 2 refer to the respective amount of -electrons participating in
the reaction. This reaction is thermally allowed. An example is the reaction of butadiene with
maleic anhydride. In a stereospecific manner, a bicyclic product is formed, that can be
transformed to the fungicide Captan, used in agriculture.
Obviously, the transition state has 6 -electrons, and no phase dislocation. According to the
aromaticity principle, this is indeed a thermally allowed process, as empirically found.

O

O

H

O

+

H

O

N

O

S
CCl3

O

H

H

O

O

Captan

Hückel type aromatic TS (6electrons)

A second approach uses the frontier orbital method. For the reaction of butadiene with ethene
one can involve either the HOMO (butadiene)/LUMO (ethene) interaction or the HOMO

(ethene)/LUMO (butadiene) interaction. Both interactions are favourable, in other words the
frontier orbitals have compatible symmetry. It is said that the reaction is symmetry- allowed.

14


HOMO butadiene
(

LUMO butadiene
(

m (A) ; C2 (S)

m (S) ; C2 (A)

LUMO ethene
(

HOMO ethene
(

m (A) ; C2 (S)

m (S) ; C2 (A)

Finally, according to the Woodward-Hoffmann rules, the Diels-Alder reaction is a suprasupra 4s + 2s process, and hence allowed. Supra-supra means that the bonds are broken or
formed on the same side, which explains the cis-stereospecificity.

In many cases it is possible to form two isomers as a result of the Diels-Alder reaction,

namely an exo- and an endo-isomer. In many cases, the latter isomer is preponderantly or
even specifically formed, even if this is the isomer that suffers the most from steric hindrance.
The names endo and exo refer to the spatial relation between the groups on the dienophile and
the newly formed bond on the diene. When these groups are on the same side, this is the
endo-adduct, otherwise this is the exo-adduct. As an example we can consider the reaction
between cyclopentadiene and maleic anhydride, leading specifically to the endo-product. On
the other hand, the reversible Diels-Alder reaction of furan with maleic anhydride affords
mainly the exo-adduct. This is a typical example of kinetic versus thermodynamic control.

O
O

H

H

O

H

O

O
O

O

H

H

H
O

O

exo-adduct
(not formed)

endo-adduct

O

O

O
O

H

H

slow

H

fast

O
O

O

O
H

H

O

H

exo-adduct
thermodynamic product

O

O

furan
(aromatic)

endo-adduct
kinetic product

15

O

The endo-specificity for irreversible reactions can be explained by frontier orbital theory. For
instance, during the formation of the endo-product from the dimerisation of cyclopentadiene
we can consider, next to the expected favourable interactions between the frontier orbitals, the
occurrence of secondary interactions (separately shown) that have bonding character and thus
favour the reaction kinetically. Obviously, the secondary interactions do not lead to bond
formation but they will lower the energy of the transition state (and hence the activation
energy). These secondary interactions are not possible during the formation of the exo-adduct.

HOMO
cyclopentadiene
(reacts as diene)

secondary
interactions
(bonding)

LUMO
cyclopentadiene
("dienophile")

Another possibility to form isomers as a result of the Diels-Alder reaction occurs if both
reaction partners, diene and dienophile, are nonsymmetrically substituted. In this case there is
the possibility of two regioisomers that differ in the relative place of the substituents of the
product obtained. In practise, often regioselectivity is observed: one of the possible
regioisomers is preferentially formed. This is a result of the electronic complementarity of the
reagents. The most common situation is the one where an electron rich diene is combined
with an electron rich dienophile. Because the reagents are non-symmetrical, some of the
orbital coefficients will be larger than others. The size of these orbital coefficients can be
calculated but often a logic is followed that can be derived from well-known considerations of
resonance or chemical reactivity. As an example we look at the reaction between methyl

acrylate (methyl propenoate) and 4-methyl-1,3-pentadiene. Methyl acrylate is the dienophile
and thus will react via a LUMO (*) with low energy. The orbital with the largest coefficient
is located on the -carbon atom. This corresponds to the most reactive (most electrophilic)
site. The 4-methylpentadiene is more electron rich than butadiene by hyperconjugation
involving the two methyl groups. The HOMO (2) has a significantly larger coefficient on the
unsubstituted end of the diene. Again, this is the most reactive (most nucleophilic) site.

16


Since both reaction partners are nonsymmetrical, the reaction itself loses it symmetry. This
reaction stays concerted but in de transition state the formation of the bond between the
termini with the larger orbital coefficients is much further advanced in comparison with the
other σ-bond. This is an explanation of the unexpected regioselectivity, forming the 1,2disubstituted product with the most steric hindrance.
The two remaining termini can bear a stabilised, complementary partial charge in the
transition state, without loss of stereoselectivity in the final product (where appropriate). This
is a so-called asynchronous process: the formation of the bonds does not occur at the same
moment although the reaction stays concerted.


CH3

HOMO
4-methyl-1,3-pentadiene
CH3

CH3

CH3

OCH3

OCH3

LUMO
methyl acrylaat

O

O

transition state

1,2 ("ortho")

Another possibility is the reaction of 2-methoxybutadiene with acrylonitrile (propenenitrile).
In this case the substituents are in 1,4-relation to each other in the cyclohexene formed. This
again is a consequence of the orbital coefficients. It is said that the Diels-Alder reaction
orients “ortho” and “para”.

H3CO

HOMO
2-methoxy-1,3-butadiene

H3CO

OCH3

OCH3

LUMO
acrylonitrile

O

O

transition state

1,4 ("para")

17


Lewis acids in combination with dienophiles further lower the LUMO in energy by
complexation with the heteroatoms present, and also the orbital coefficient (at the  position
in relation to this heteroatom) will increase. Thus, the reactions will be faster and with higher
regioselectivity. Isoprene (2-methylbutadiene) reacts with methyl vinyl ketone (1-propen-3one) only after heating in toluene in a closed reaction vessel, and an isomer mixture of the
1,4- and 1,3-substituted product is formed in a 71:29 ratio. After addition of SnCl4.5H2O the
reaction becomes possible at 0°C, and the ratio improves to 93:7.

H3C

H3C

+

and

H3C
CH3
O

CH3

O

toluene, 120°C

71

SnCl4.5H2O, 0°C

93

O

:

29

:

7

1.3.2. [2+2]Cycloadditions

After irradiation of alkenes with UV-light, cyclobutane derivates may form. This is a
pericyclic reaction that normally does not occur when alkenes are heated (normally alkenes

polymerize on heating). Thus, the photochemical dimerisation reaction is allowed.
Application of the aromaticity rule shows that a supra-supra approach implies a Hückel antiaromatic system, thus thermally the reaction is forbidden. An alternative approach, supraantara in which the two alkenes approach in a perpendicular fashion in the transition state,
leads to an aromatic 4-electron Möbius system (one phase dislocation) but this is difficult to
realise by ring strain and steric hindrance of the substituents on the alkenes in the transition
state.

18


+

cyclobutane

Hückel anti-aromatic

=
Möbius aromatic

perpendicular approach

1 phase dislocation
4 electrons

Via the frontier orbital approach it is possible to see that the photochemical reaction is indeed
allowed. After irradiation and absorption of a photon an electron is promoted from the - to
the *-orbital, which now is the HOMO. If we combine this excited molecule with a molecule
in the ground state, the symmetry of the frontier orbitals identical. For two molecules in the
ground state, the symmetry of the frontier orbitals is opposite and this reaction is forbidden.

HOMO alkene

, m

HOMO alkene (excited state)
*, C2

LUMO alkene
*, C2

LUMO alkene
*, C2

thermally : symmetry-forbidded

photochemically : symmetry-allowed

Ketenes or other electron poor cumulenes (such as isocyanates RN=C=O) will smoothly
undergo thermal [2+2] cycloadditions with electron rich alkenes. The perpendicular approach
of the two reagents gives a situation in which the frontier orbitals (LUMO of ketene, HOMO
of alkene) are stabilised by the p-orbital on the central carbon, that is part of the C=O bond.
The latter orbital is perpendicular to the p-orbitals of the C=C bond and therefore is
overlapping with the HOMO of the alkene. Moreover, the central carbon atom of the ketene is

19


sp-hybridised and unsubstituted is, minimising the steric interactions in the transition state
and the product. An example is the cycloaddition of dichloroketene with cyclopentadiene.
Notably, the [2+2]-cycloaddition takes preference over the [4+2]-cycloaddition !

H

O

+

O

C
Cl
C
Cl

Cl

Cl

H

p orbital
LUMO

antibonding
interaction

bonding (stabilising) interactions

HOMO
frontier orbitals of the cycloaddition of a ketene and an alkene

1.3.3 Other cycloadditions

There exists a large variety of higher cycloadditions, to which the principles discussed earlier
can be applied. For instance, cyclopentadiene reacts with tropone (cycloheptatrienone) in a
thermally allowed 6s + 4s addition. The exo-adduct is formed preferentially because the
secondary interactions during the formation of the endo-isomer are antibonding.
O

O

O

+
+

EXO

ENDO

(main product)

antibonding interactions
X

X
bonding interactions

O

20

1,3-Dipolar cycloaddition reactions occur via molecules that are similar to the allyl anion,
thus they have 4 -electrons and they can react with a suitable unsaturated compound, then
named a dipolarophile (mostly alkenes or alkynes). The mechanism bears analogy to the
Diels-Alder cycloaddition. Well-known 1,3-dipoles are diazoalkanes, azides, and ozone.
Ozonolysis is a 1,3-dipolar cycloaddition which occurs via a 1,2,3-trioxolane, that undergoes
a cycloreversion (the opposite of a cycloaddition) to a new, very reactive 1,3-dipole, a
carbonyl oxide, and a ketone. Alternative 1,3-dipolar cycloaddition affords the ozonide (a
1,2,4-trioxolane), that can be reduced, for instance with dimethyl sulfide, to aldehydes (or
ketones for tri- or tetra substituted alkenes).

R
N

N

N

C

R

N

O

N

O

O

R

R
N

N

N

C

R

N

O

N

O

azides

ozone

diazoalkanes

O
O

O

R

O
O

O
O

O

O

O

O

O

1,3-DC

O

1,3-DC
isoozonide

carbonyl oxide
1,3-dipole

ozonide
DMS

-DMSO

O

+

O

1.3.4 The ene reaction
This reaction was discovered by Alder and named the “ene”-reaction to distinguish it from the
“diene”-reaction reported earlier by Diels and himself. From the name we can guess that this
is a reaction involving alkenes. It is possible to look at this reaction as an analog of the DielsAlder reaction in which a C-H -bond replaces a double bond of the diene. In this reaction, no
ring is formed, but rather a new C-C bond, and a hydrogen atom is relocated through space.

21


As concerns the orbitals, there are clear differences between the ene reaction and the DielsAlder reaction. The C-H bond is parallel to the p-orbitals of the (alk)ene, in such a way that
after the reaction a new double bond may be formed. The two molecules approach each other
in parallel planes. The ene has two components, a 2- and a 2-component. Next to these we
have a 2-component of the alkene (anhydride). The latter is in most cases an electron poor
alkene, reacting via its LUMO with the HOMO of the 2- and 2-components of the ene.
These electron poor reagents are called enophiles.
The three components are all of the (4q+2)s type, and application of the Woodward-Hoffmann
rules confirms that the reaction is thermally allowed. The aromaticity rule (no phase
dislocation, 6 electrons) and the frontier orbital theory are also in agreement with this.

O

O

H

O

O

O

H

O

H

O

O

Diels-Alder reaction

O
H

H
O

O

H

O

Alder ene reaction

HOMO ()

"ene"

bonding
HOMO ()

H
O

bonding

LUMO (*)
O

(electron poor)
alkene
enophile

O

A carbonyl group is a good enophile and the corresponding reactions with alkenes are called
carbonyl-ene reactions. Lewis acids will further increase the reactivity of the carbonyl group.
An example is the intramolecular carbonyl ene reaction of (R)-citronellal, a terpene
compound. This reaction is catalysed by the Lewis acid ZnBr2, which affords isopulegol, that
by reduction can be transformed into (-)-menthol. The stereochemistry of the carbonyl ene
reaction is explained by the occurrence of a trans-decaline transition state, in which the larger
substituents (methyl, hydroxy, isopropenyl) assume an equatorial position. Although menthol
is found in Nature, most of the commercial menthol is prepared in this way.

22


H

O

H2/Ni

ZnBr2
OH

OH

(R)-citronellal

(-)menthol

isopulegol

Me

H
O
ZnBr2

transition state :
trans-decaline system

H
H
Me

1.3.5 Cheletropic reactions
These are cycloaddition reactions in which two new σ-bonds are created on the same atom. A
non-bonding orbital (named ) that participates in these reactions can form bonds via one
lobe (suprafacially) or via both lobes (antarafacially). The reaction of an alkene with a sp2hybridised carbene (see later in this text) will occur via a non-linear approach. The linear
approach is not allowed for reasons of (orbital) symmetry. Further along the reaction course,
the CH2-groep will turn to minimise the strain in the final product (Skell mechanism).
According to the Woodward-Hoffmann rules, this is an allowed 2s + 2a-process, and the
aromaticity principle allows us to see the TS as a Möbius 4-system. More applications of
this reaction follow in the part on carbenes and nitrenes.

HOMO m
H

H

H

2s

+

H

H

2a

non-linear approach
allowed

2s

+

H

LUMO
H

LUMO m
carbene

HOMO
carbene C2

H

2s

linear approach
forbidden

Woodward-Hoffmann approach

Frontier orbital approach

23

C2


The addition of SO2 to dienes can be used to prepare sulfolenes.

This cheletropic

cycloaddition occurs via a linear approach. The SO2 molecule is electron poor and thus reacts
via its LUMO, which is analogous to that of the allyl anion. At higher temperatures, the
equilibrium is shifted from the sulfolenes to the dienes and SO2, as a consequence of the
increasing effect of the entropy factor. This is an extrusion reaction, and can be used as a
possible synthetic route towards substituted dienes. Thus, the diene is protected first as a
sulfolene, and later synthetic transformations can be carried out without interference of the
chemically labile diene system. In the last step, the diene is released by heating. In the
example below, sulfolene is transformed in the anion (well stabilised by the sulfone function)
and alkylated with 6-bromo-1-hexene. Thermolysis yields the substituted butadiene, which
will undergo an intramolecular cycloaddition (via a chair-type conformation) to a trans-fused
bicyclic system.

Base

+

SO2

SO2

SO2

6-bromohexene
HOMO

LUMO

thermolysis
180°C
Diels-Alder reaction
H

H

1.4 Sigmatropic rearrangements
In a [i,j]-sigmatropic rearrangement, a group migrates within a -system, in which the double
bonds shift during the migration. The number i refers to the (carbon) atom of the migrating
group, and j is the number if the migration terminus. The two atoms that form the original bond are given number 1. The total amount of - and -bonds does not change during a

24