Substituent effects on the reversibility of furan - Maleimide cycloadditions

Article abstract of DOI:10.1021/jo201606z

The effects of furan and maleimide substitution on the dynamic reversibility of their Diels - Alder reactivity have been investigated computationally and by ¹H NMR spectroscopy. Furan and furan derivatives bearing methoxy, methyl, or formyl groups at their 2- or 3-positions were investigated with maleimide and maleimide derivatives bearing N-methyl, N-allyl, and N-phenyl substituents. Computational predictions indicate that electronic and regiochemical effects of furan substitution significantly influence their Diels - Alder reactivity with maleimide, with reaction free energies of exo adduct formation ranging from ΔG = - 9.4 to 0.9 kcal/mol and transition state barriers to exo adduct formation ranging from ΔG? = 18.9 to 25.6 kcal/mol. Much less variation was observed for the reactivity of N-substituted maleimide derivatives and furan, with reaction and transition state free energies each falling within a range of 1.1 kcal/mol. Dynamic exchange experiments monitored by ¹H NMR spectroscopy support computational predictions. The results indicate the reactivity and reversibility of furan - maleimide cycloadditions can be tuned significantly through the addition of appropriate substituents and have implications in the use of furan and maleimide derivatives in the construction of thermally responsive organic materials.

Full text of DOI:10.1021/jo201606z

ARTICLE
pubs.acs.org/joc
CS uy bc l so t ai tdu de int ito En f sf ects on the Reversibility of FuranꢀMaleimide
Robert C. Boutelle and Brian H. Northrop*
Department of Chemistry, Wesleyan University, Middletown, Connecticut 06459, United States
S Supporting Information
b
ABSTRACT: The effects of furan and maleimide substitution
on the dynamic reversibility of their DielsꢀAlder reactivity have
1
been investigated computationally and by H NMR spectro-
scopy. Furan and furan derivatives bearing methoxy, methyl, or
formyl groups at their 2- or 3-positions were investigated with
maleimide and maleimide derivatives bearing N-methyl, N-allyl, and N-phenyl substituents. Computational predictions indicate that
electronic and regiochemical effects of furan substitution significantly influence their DielsꢀAlder reactivity with maleimide, with
reaction free energies of exo adduct formation ranging from ΔG = ꢀ9.4 to 0.9 kcal/mol and transition state barriers to exo adduct
q
formation ranging from ΔG = 18.9 to 25.6 kcal/mol. Much less variation was observed for the reactivity of N-substituted maleimide
derivatives and furan, with reaction and transition state free energies each falling within a range of 1.1 kcal/mol. Dynamic exchange
1
experiments monitored by H NMR spectroscopy support computational predictions. The results indicate the reactivity and
reversibility of furanꢀmaleimide cycloadditions can be tuned significantly through the addition of appropriate substituents and have
implications in the use of furan and maleimide derivatives in the construction of thermally responsive organic materials.
1
4
13
’
INTRODUCTION
from colloids and coordination complexes to mechanically
16 17 18
interlocked molecules such as rotaxanes, catenanes, suitanes,
Borromean rings, and Solomon knots. More recently sev-
eral research groups
of boronate esters to assemble a variety of covalent organic
frameworks (COFs) with applications in gas uptake and
storage, organic photovoltaics, and catalysis. Severin and
Nitschke have demonstrated the simultaneous use of two or
more dynamic covalent reactions in the synthesis of macrocycles
and capsules.
Dynamically reversible covalent bond forming reactions have
1
9
20
been the subject of increasing interest over the past two decades,
largely because of their relevance to biological chemistry, mo-
lecular and macromolecular materials, and supramolecular chem-
istry. Dynamic covalent reactions proceed via relatively low
activation barriers and form products that are only slightly
exergonic relative to starting reagents, thus allowing the reverse
reaction to also take place under appropriate conditions. The
potential for both forward and reverse reactivity allows for the
2
1ꢀ24
have used the reversible condensation
2
5
2
6
1
,2
It has been known for over 50 years that some DielsꢀAlder
construction of dynamic combinatorial libraries (DCLs) and
forms the basis of dynamic covalent chemistry (DCC),
27
3
ꢀ5
reactions can be made reversible. The application of reversible
DielsꢀAlder chemistry to the dynamic covalent assembly of
wherein mixtures of multiple components undergo continual
exchange under thermodynamic equilibrium conditions. Exam-
ples of dynamic covalent reactions include imine condensations,
aldol exchange, transesterifications, disulfide formation, boronic
ester condensations, and alkene metathesis. Dynamic covalent
exchange provides a means of adaptive proofreading as kinetic
intermediates are cycled through en route to the most thermo-
dynamically stable product (or products). Similar concepts of
dynamic reversibility have been exploited for over half a century
2
8ꢀ40
organic materials, however, has only recently been explored.
This comes despite the fact that DielsꢀAlder reactions have the
desirable property of being self-contained: i.e., all atoms present in
the starting diene and dienophile are present in the resulting
adduct. The equilibrium position of reversible DielsꢀAlder reac-
tions, therefore, responds to temperature, solvent, and concentra-
tion but is not influenced by ancillary molecules such as water in
the case of imine and boronic ester condensations or transition-
metal catalysts in the case of alkene metathesis. The number of
DielsꢀAlder reactions that undergo dynamically reversible adduct
formation under mild conditions is, however, limited and there is
increasing interest in expanding the set of dynamic covalent
6
ꢀ10
in the context of noncovalent self-assembly.
Dynamic cova-
lent chemistry has the advantage of combining the reversibility of
noncovalent self-assembly with the strength of covalent bond
formation. This adaptability, ease of synthesis, and product
stability has made dynamic covalent chemistry particularly well
suited to the construction of organic materials.
Many early applications of DCC focused on the formation of
synthetic receptors utilizing imine, ester, disulfide, and acetal
chemistries. Template-directed dynamic assembly
41ꢀ44
DielsꢀAlder reactions.
Over the past 10 years several groups have explored the use of
reversible DielsꢀAlder adduct formation in the design and
1
1
12ꢀ15
has
Received: August 1, 2011
Published: August 26, 2011
enabled the efficient construction of various materials ranging
r 2011 American Chemical Society
7994
dx.doi.org/10.1021/jo201606z
|J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
Scheme 1. Parent [4 + 2] Cycloaddition of Unsubstituted
Furan with Maleimide To Give the Kinetic Endo and
Thermodynamic Exo DielsꢀAlder Adducts
In the same study the authors investigated the cycloaddition of
isobenzofuran with N-methylmaleimide (reaction 3, Scheme 2).
Adduct formation in reaction 3 involves the formation of a stable
aromatic product and is therefore considerably more favorable.
The thermodynamic equilibrium for the reaction of isobenzofur-
an and N-methylmaleimide therefore requires heating at 132 °C
in chlorobenzene in order to observe reversibility. The authors
determined a ΔΔGendo-exo value of 2.4 kcal/mol. Computational
methods presented herein were benchmarked against these
experimental studies in order to test their appropriateness, and
the results are shown in Table 1. Density functional methods at
the B3LYP/6-311 g(d,p) level deviated significantly from experi-
mental values in the gas phase in solvent models for both
acetonitrile and chlorobenzene. This observation is consistent
with prior computational research that has shown B3LYP
synthesis of new organic materials. The majority of such studies
utilize the reactivity of electron-rich furan derivatives with
electron-poor maleimide derivatives. The [4 + 2] cycloaddition
of furan and maleimide can be accomplished at or slightly above
room temperature, while the retrocyclization is accomplished at
elevated temperatures (Scheme 1). Wudl and co-workers have
developed transparent organic polymers that, when cracked, are
thermally remendable on account of reversible furanꢀmaleimide
51
methods to be accurate for pericyclic reactions involving
45
hydrocarbons but unreliable for pericyclic reactions involving
heteroatoms. More computationally intensive MP2 and CBS
methods, on the other hand, proved significantly more accurate.
The best agreements with experimental results were achieved
utilizing single-point electronic energies obtained at the MP2/
2
8
cycloadditions. Similar demonstrations of the utility of furanꢀ
maleimide cycloadditions in the preparation of molecular mate-
rials include thermally reversible dendrimers and dendronized
2
9ꢀ31
32
33
polymers,
surfactants, epoxy resins, cross-linked poly-
6
-311+G(d,p) level with vibrational frequency analysis carried out
3
4ꢀ36
37 38,39
mer networks,
and the aggregation and manipulation
at the CBS-QB3 level. Transition states were optimized similarly,
of gold nanoparticles. Each of the preceding examples involve
furan compounds that are mono- or disubstituted with alkyl,
27,31,45
Table 1. Experimental and Calculated Reaction (ΔG) and
ether, ester, or amine substituents. It is well-known
that
q
Transition State (ΔG ) Free Energies for DielsꢀAlder
the substitution of furans greatly effects their reactivity with
dienophiles, such as maleimide. Despite this sensitivity to sub-
stitution, and the increasing application of furanꢀmaleimide
cycloadditions to materials development, there have yet to be any
systematic investigations of the effects of furan and maleimide
substitution on the thermodynamics of their DielsꢀAlder reactivity.
We report herein computational and spectroscopic investiga-
tions of the electronic and regiochemical effects of furan sub-
stitution on the reversibility of furanꢀmaleimide cycloaddition
reactions and discuss their implications on the use of dynamic
furanꢀmaleimide cycloadditions in materials synthesis. The
results show that, with the appropriate substituents, furanꢀ
maleimide cycloadditions can be tuned along a scale ranging
from exergonic to dynamically reversible to endergonic.
Cycloadditions between Furan and N-Methylmaleimide
(
Reaction 2) and between Isobenzofuran and
N-Methylmaleimide (Reaction 3) Used To
Benchmark Computational Methods
a
Reaction 2
a
exptl
computational
CH
3
3
CN/313 K in vacuo/298 K CH CN/298 K CH
3
CN/313 K
q
ΔG
q
ΔG
q
ΔG
q
ΔG
ΔG
ΔG
ΔG
ΔG
endo ꢀ1.9
exo ꢀ3.7
24.9
25.2
ꢀ2.0
ꢀ4.1
22.4
22.6
ꢀ1.5
ꢀ4.5
22.5
22.7
ꢀ1.3
ꢀ4.4
23.1
23.4
Reaction 3
’
RESULTS AND DISCUSSION
exptl
computational
DielsꢀAlder cycloaddition reactions of furan and maleimide
46ꢀ48
have been studied both experimentally
and computat-
PhCl/405 K
in vacuo/298 K
PhCl/298 K
PhCl/405 K
4
9
ionally. Furanꢀmaleimide adduct formation proceeds to give
exo and endo products, with the exo species being the thermo-
q
q
q
q
ΔG
ΔG
ΔG
ΔG
ΔG
ΔG
ΔG
ΔG
50
dynamic product and endo the kinetic. Rickborn et al. report
endo ꢀ17.1 18.2
exo ꢀ19.5 19.4
ꢀ25.7
9.2 ꢀ19.6 14.3
that, in the case of furan reacting with N-methylmaleimide
at 40 °C in acetonitrile solution, the exo isomer is more
stable than the endo by 1.8 kcal/mol (reaction 2, Scheme 2).
ꢀ26.5 10.4 ꢀ20.4 15.2
a
All values are in kcal/mol.
Scheme 2. Diels-Alder Cycloaddition Reactions between Furan and N-Methylmaleimide (Reaction 2) and between
Isobenzofuran and N-Methylmaleimide (Reaction 3), the Dynamic Reversibility of Which Have Been Studied Experimentally
7
995
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
Figure 1. Chemical structures of the substituted furan (a) and maleimide (b) derivatives investigated in the present study.
though using less computationally intensive MP2/6-31+G(d)
electronic energies. Reaction and transition state free energies
were obtained in the gas phase at 298 K, solvated in either
Table 2. Calculated Reaction Enthalpies (ΔH), Transition
q
State Enthalpies (ΔH ), Reaction Free Energies (ΔG), and
q
Transition State Free Energies (ΔG ) (kcal/mol) for
CH CN or PhCl at 298 K, and solvated at experimental tem-
FuranꢀMaleimide DielsꢀAlder Cycloadditions
3
peratures (313 or 405 K). Computations accurately predict the
endo isomer to be favored kinetically and the exo isomer to be
favored thermodynamically in each case for reactions 2 and 3.
The closest agreement between calculated and experimental
reaction and transition state free energies for the reaction of
furan with N-methylmaleimide, reaction 2, was obtained when
including solvent and correcting for the experimental tempera-
ture. Calculated reaction free energies in this case are within
q
q
q
furan maleimide product isomer ΔH ΔH ΔG ΔG ΔGMeCN ΔG MeCN
0
.7 kcal/mol of experimental values. Calculations accurately
reproduce the experimentally observed difference in transition
1
5
5
5
5
5
5
5
6
7
8
9
endo ꢀ16.0 9.4 ꢀ1.2 23.1 ꢀ1.1
exo ꢀ18.1 9.5 ꢀ3.3 23.2 ꢀ3.6
endo ꢀ21.9 6.6 ꢀ5.8 21.3 ꢀ5.9
exo ꢀ22.6 6.3 ꢀ6.4 21.3 ꢀ6.0
endo ꢀ22.3 5.8 ꢀ7.4 19.6 ꢀ7.5
exo ꢀ24.2 6.4 ꢀ9.4 20.0 ꢀ9.4
endo ꢀ18.0 7.2 ꢀ2.7 21.4 ꢀ2.4
exo ꢀ19.3 7.6 ꢀ4.0 21.6 ꢀ4.0
endo ꢀ18.9 6.8 ꢀ3.9 20.8 ꢀ3.6
exo ꢀ20.3 7.7 ꢀ5.4 21.3 ꢀ5.8
23.0
23.0
20.5
20.8
18.9
19.3
21.3
21.6
20.9
21.0
25.6
25.4
24.8
24.3
22.5
22.7
21.9
22.1
21.8
21.6
q
state free energies of ΔΔG
= 0.3 kcal/mol for reaction 2,
2a
2b
10
11
12
13
14
15
16
17
18
endo-exo
while absolute predictions of the exo and endo transition state
free energies are within 1.2 kcal/mol. Likewise, correcting for
solvent and experimental temperatures gave the best agreement
for the reaction of isobenzofuran with N-methylmaleimide
3a
3b
4a
4b
1
1
1
(reaction 3). Calculated reaction free energies for this reaction
deviate from experimental values to a greater extent in absolute
terms (0.9ꢀ2.5 kcal/mol), though they still reproduce experi-
mental trends. With these benchmarking studies at hand, the
effects of furan and maleimide substitution on their DielsꢀAlder
reactivity were then investigated.
Figure 1 shows the series of substituted furan and maleimide
compounds chosen for this study. Furan and furan derivatives
with a methoxy, formyl, and methyl substituent at either the 2- or
endo ꢀ11.4 12.1 3.2 25.9
exo ꢀ14.4 11.3 0.3 25.1
endo ꢀ14.1 10.4 0.5 24.1
3.5
0.9
1.0
exo ꢀ18.0 10.4 ꢀ3.4 24.0 ꢀ3.7
endo ꢀ17.3 8.7 ꢀ3.0 22.4 ꢀ1.5
exo ꢀ19.2 9.0 ꢀ4.5 22.6 ꢀ4.5
endo ꢀ16.5 8.1 ꢀ1.7 21.9 ꢀ1.4
exo ꢀ18.4 8.7 ꢀ3.6 22.5 ꢀ3.8
endo ꢀ16.7 8.0 ꢀ2.0 21.9 ꢀ2.4
exo ꢀ18.5 8.7 ꢀ3.8 22.3 ꢀ4.8
3
-position were investigated. This particular series was chosen for
their widely varying electronic character (electron donating,
withdrawing, and alkyl) as well as the commercial availability
of most compounds in the series. In the current study only
monosubstituted furan derivatives were investigated. Maleimide
and maleimide derivatives with methyl, vinyl, and phenyl sub-
stituents were also studied.
proceed with ΔG = ꢀ6.0 kcal/mol. This value is 2.4 kcal/mol
more favorable than the parent reaction with unsubstituted furan.
The effect is even greater in the case of 3-methoxyfuran (2b),
where adduct formation is predicted to be favored by ꢀ9.4 kcal/
mol, which is 5.8 kcal/mol more favorable than the parent
reaction with unsubstituted furan. In addition to increasing the
reaction free energy, methoxy substitution lowers the transition
state free energy barrier to adduct formation relative to unsub-
stituted furan (1) by 1.5ꢀ2.2 kcal/mol. While methoxy substitu-
tion is predicted to both increase the exergonicity of and lower
the barrier to adduct formation, computations predict the greater
effect is seen in the overall reaction free energy: i.e. the effect is
Electronic Effects. Table 2 summarizes the calculated reaction
and transition state energies and free energies for DielsꢀAlder
cycloadditions between furan (1) and maleimide (5) as well as
between furan derivatives 2a,bꢀ4a,b and maleimide derivatives
6
ꢀ8. It is well-known that DielsꢀAlder cycloadditions involving
electron-rich dienes and electron-poor dienophiles proceed
more favorably.
2
7,45
This trend is borne out across the series of
compounds investigated computationally. The effects of adding
alkoxy, alkyl, and aldehyde sutstibutents to furan, whether at the
2
- or 3-position, generally follow what would be expected, given
their Hammett parameters. The greatest increase in exergonicity
of adduct formation is seen with the most electron donating
methoxy substituent. In the case of 2-methoxyfuran (2a),
formation of the exo adduct with maleimide is predicted to
q
more pronounced for ΔG than for ΔG . These results predict
that the addition of strongly electron donating substituents to
furan should decrease their utility as partners with maleimide in
7
996
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
dynamic covalent chemical applications. The significantly lower
reaction free energies result in high barriers to retrocyclization:
Table 3. Calculated Reaction Enthalpies for the Reactions
between Furan and Dimethyl Ether, Ethane, and
e.g., the cycloaddition of 3-methoxyfuran (2b) with maleimide
Acetaldehyde To Produce either a 2- or 3-Substituted
Furan and Methane
q
a
(
5) to give the product 11-exo has ΔG = 19.3 kcal/mol, whereas
its retrocyclization requires ΔG = 28.7 kcal/mol.
q
Methyl substitution is also predicted to increase the favor-
ability of adduct formation, though to a lesser extent. Reacting
2
-methylfuran (3a) with maleimide (5) lowers the free energy of
endo and exo product formation by 1.3 and 0.4 kcal/mol,
respectively. This increase in favorability increases slightly when
3
-methylfuran (3b) is used as the starting diene, with endo and
exo product formation predicted to be more favorable by 2.5
and 2.2 kcal/mol, respectively, relative to unsubstituted furan.
Calculated free energy barriers to the formation of Dielsꢀ
Alder adducts between methylfurans and maleimide are lower
than the corresponding barriers for the parent adducts by
a
Enthalpies are given in kcal/mol.
1
2
.4ꢀ1.7 kcal/mol, in the case of 2-methylfuran, and 2.0ꢀ
.1 kcal/mol, in the case of 3-methylfuran. It is therefore
predicted that the increased favorability of adduct formation is,
within error, balanced out by a decrease in the free energy barrier
to adduct formation.
Aldehyde substitution is predicted to increase the free energy
barriers to adduct formation by 1.3ꢀ2.6 kcal/mol relative to an
unsubstituted furan. These results are generally supportive of the
precedent that electron-poor dienes, such as 4a, react less
favorably with electron-poor dienophiles than electron-rich
dienes, such as 2b. It is surprising, however, that the electron-
poor diene furan 4b is predicted to react as favorably with
maleimide 5 as is the unsubstituted and more electron rich
parent furan 1. This notable difference in predicted reactivity
necessitated a more thorough investigation of the regiochemical
effects of furan substitution.
Regiochemical Effects. The results of Table 2 indicate that
furan substitution at the 3-position consistently leads to more
negative reaction free energies and lower transition state barriers
than substitution at the 2-position. This effect is present across
the series of substituted furans 2ꢀ4 and is most pronounced in
the case of aldehyde substitution. Isodesmic reactions were used
to determine the origin of this increased favorability of adduct
formation involving 3-substituted furans.
As can be seen in Table 3, isodesmic reactions corresponding
to substitution of furan relative to methyl are exothermic in all
cases, though they are more exothermic for substitution at the
2-position than at the 3-position. The 2-position of unsubstituted
furan is more electropositive than the 3-position, which explains
why substitution at the 2-position is most favored in the case of
methoxy (reaction 4a, ΔH = ꢀ9.5 kcal/mol), the strongest
donating substituent. This also explains the differences in favor-
ability between substitution at the 2-position versus 3-position
In effect, alkyl substitution provides a means of developing
more complex and synthetically useful furan dienes to be used as
partners with maleimide dienophiles in dynamic covalent chemi-
cal applications without disrupting the dynamic reversibility of
DielsꢀAlder adduct formation. These results are especially
important given the increasing number of examples wherein
furan dienes, substituted at either the 2-position or the 3-position
with either alkyl groups or alkyl ethers, have been used in
conjunction with maleimide derivatives for the dynamic covalent
synthesis of organic materials such as surfactants, polymers, and
31
dendrimers. Indeed, Kakkar et al. have experimentally observed
the differences in reactivity of furans substituted at their 2-posi-
tion with either alkyl or ester functionalities: 2-alkyl furans
underwent reversible DielsꢀAlder reactions with maleimides
to give thermoresponsive dendrimers, whereas direct substitu-
tion with electron-poor ester functionalities resulted in furans
that were unreactive toward maleimides. These experimental
observations are in agreement with the computational predic-
tions presented herein. Furthermore, it ispredicted that while both
2
- and 3-methylfuran will react reversibly with maleimide at high
temperatures, alkyl substitution at the 3-position will result in a
more stable, more robust DielsꢀAlder adduct on return to
ambient temperatures. Such subtle distinctions are important for
the development of, for example, thermally remendable polymers.
Interestingly, and somewhat counterintuitively, computations
predict the addition of an electron-withdrawing aldehyde sub-
tituent to furan results in either endergonic or exergonic Dielsꢀ
Alder reactivity with maleimide, depending on the regiochem-
istry of furan substitution. Reacting maleimide with 2-furalde-
hyde (4a) is predicted to be endergonic by 3.5 and 0.9 kcal/mol
for endo and exo isomers, respectively. The same reaction with
(ΔΔH(
) across the three substituents investigated,
3-sub)ꢀ(2-sub)
which is smallest for the most withdrawing aldehyde substituent
(reactions 5cꢀ4c, 0.8 kcal/mol) and greatest for the most
donating methoxy (reactions 5aꢀ4a, 2.9 kcal/mol). The overall
trend in Table 2 indicates that the 3-substituted regioisomer of a
given substituent is destabilized relative to the 2-substituted
regioisomer. This difference ultimately plays an important role
in the DielsꢀAlder reactivity of substituted furans with maleimide.
Isodesmic reactions in Table 4 were used to evaluate the
effects of substitution on the furanꢀmaleimide DielsꢀAlder
adduct. Adduct formation has the two prominent effects of (i)
alkylation at the 2- and 5-positions of the furan moiety and (ii)
changing the π-system of the furan moiety from a 2,4-diene to a
3-alkene. These changes in alkylation and conjugation alter the
trends in substituent favorability, as can be seen by comparing
related isodesmic reactions in Tables 3 and 4. Methoxy substitution,
3
-furaldehyde (4b), on the other hand, is predicted to be endergonic
by 1.0 kcal/mol in the case of the endo adduct and exergonic by
.7 kcal/mol in the case of exo adduct formation. This significant
3
distinction again highlights the notable difference in reactivity of
furans substituted at the 2-position and those substituted at the
3
-position. Across the series of furan derivatives studied, those
substituted at the 3-position react with maleimide to give exo
DielsꢀAlder adducts that are 1.8ꢀ4.8 kcal/mol more favored
than their corresponding 2-substituted isomers. This difference
is greatest in the case of 2-furaldehyde versus 3-furaldehyde.
7
997
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
Table 4. Calculated Reaction Enthalpies for the Reactions
between FuranꢀMaleimide Diels-Alder Adduct 9 and
Dimethyl Ether, Ethane, and Acetaldehyde To Produce
a
the Indicated 2- or 3-Substituted Adduct and Methane
Figure 2. Plot comparing calculated reaction free energies (ΔG, x axis)
and transition state free energies (ΔG , y axis) in acetonitrile for
q
a
Enthalpies are given in kcal/mol.
furanꢀmaleimide DielsꢀAlder cycloadditions investigated in this study.
Endo isomers are symbolized by diamonds, while exo isomers are
symbolized by squares. Adducts containing substituted furans are shown
in either red (endo) or blue (exo) and labeled according to their
substitution. Adducts containing substituted maleimides are shown in
gray. Endo and exo adducts of the unsubstituted furanꢀmaleimide
DielsꢀAlder adduct are shown in green.
for example, is 4.6ꢀ6.4 kcal/mol more favorable in reactions 6a and
7a than in reactions 4a and 5a. This effect appears to be primarily the
result of the change in conjugation rather than increased alkylation,
given that the increases in reaction exothermicities on going from
Table 3 to Table 4 are roughly the same whether methoxy sub-
stitution occurs at the 2-position (comparing reactions 4a and 6a) or
the 3-position (comparing reactions 5a and 7a). Methyl substitution
in reactions 6b and 7b shows only a modest increase in exothermi-
city of 1.2ꢀ2.9 kcal/mol relative to reactions 4b and 5b. For both
methoxy and methyl substituents, substitution of the DielsꢀAlder
adduct at the 2-position of the furan moiety is predicted to be more
favorable than substitution at the 3-position, a trend also observed in
Table 3.
In the case of aldehyde substition this preference is reversed:
i.e., substitution at the 3-position of the furan moiety is more
exothermic than that at the 2-position (Table 4). This role
reversal can be explained by the fact that aldehyde substitution
at the 3-position retains π-conjugation between the adduct and
the aldehyde CdO, which is lost for aldehyde substitution at the
3-furaldehyde with maleimide results in the formation of
minor amounts of adduct 15 (see the Supporting Information),
whereas all attempts to react 2-furaldehyde with maleimide to
give adduct 14 resulted in no reaction, in line with the predicted
results.
Computational results presented herein suggest that the reac-
tivity and reversibility of DielsꢀAlder reactions between substi-
tuted furans and maleimide can be tuned through the addition of
appropriate substituents. Adduct formation can range from being
strongly exergonic and thus irreversible (e.g., the reaction of
3-methoxyfuran 2b with maleimide 5) to slightly exergonic with
increased reversibility (e.g., the reaction of 2-methylfuran 3a with
maleimide 5) to endergonic (e.g., the reaction of 2-furaldehyde 4a
with maleimide 5). These results have significant implications
for the use of furanꢀmaleimide cycloadditions in the dynamic
covalent synthesis of organic materials. While methoxy substitu-
tion ispredicted to lower thebarrier toadduct formation and result
in more stable DielsꢀAlder adducts, this favorability comes at the
cost of significantly hindering reversibility.
Maleimide substitution was also investigated, and the results
are presented in Table 2. The reactivity of N-alkyl-, N-allyl-, and
N-phenyl-substituted maleimides with furan were studied. Sub-
stitution of maleimide dienophiles is predicted to have a less
pronounced effect on reaction and transition state energetics
than substitution of furan dienes. Overall, all three substituted
maleimides increase the exergonicity of DielsꢀAlder adduct
formation relative to unsubstituted maleimide by 0.2ꢀ1.3 kcal/
mol, with methyl substitution showing the greatest effect.
Maleimide substitution is also predicted to lower free energy
barriers to adduct formation by 0.3ꢀ1.4 kcal/mol. While the
effects of maldimide substitution are not as large or as variable as
the effects of furan substitution, it is notable that the predicted
decrease in transition state barriers is, within error, equal to the
decrease in reaction free energies. It can therefore be expected
that the addition of alkyl or aromatic groups to maleimide will
not hinder but, rather, will likely facilitate their use in dynamic
covalent chemical applications with furan and furan derivatives
2
4
-position. Taken together, the results presented in Tables 3 and
show that aldehyde substitution at the 2-position of furan
(
(
reaction 4c) is stabilized relative to substitution at the 3-position
reaction 5c), whereas the DielsꢀAlder adduct in reaction 6c is
destabilized relative to the DielsꢀAlder adduct in reaction 7c.
The increase in favorability of aldehyde substitution of the
DielsꢀAlder adduct at the 3-position (1.9ꢀ2.9 kcal/mol) out-
weighs the 0.8 kcal/mol destabilization of aldehyde substitution
of furan at the 3-position relative to the 2-position (Table 3). The
difference in predicted reactivity between 2-furaldehyde and
3
-furaldehyde with maleimide can therefore be attributed to
two main factors: (i) aldehyde substitution at the more electro-
positive 2-position is less favorable than at the 3-position and (ii)
the loss of π-conjugation upon DielsꢀAlder adduct formation in
the case of 2-furaldehyde further disfavors adduct formation,
whereas π-conjugation is retained in the case of 3-furaldehyde.
The enthalpic effects investigated in isodesmic reactions of
Tables 3 and 4 are able to support the predicted though counter-
intuitive reaction free energies presented in Table 2: namely, that
formation of an exo DielsꢀAlder adduct between 3-furaldehyde
and maleimide is predicted to be exothermic despite the electron-
withdrawing nature of the aldehyde substituent. In the case of
2
-furaldehyde reacting with maleimide, both exo and endo adduct
formation is endothermic and unfavorable. Indeed, reacting
7
998
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
1
Scheme 3. Dynamic Exchange Reactions Involving the Exchange of Maleimide (a) or Furan (b) Moieties Studied by H NMR
Spectroscopy
1
Figure 3. Partial H NMR spectra recorded at t = 0 h (top), t = 51 h
(middle), and t = 400 h (bottom), demonstrating maleimide/N-
methylmaleimide exchange and equilibration between adducts 9 and 16.
by slightly increasing the thermodynamic favorability of Dielsꢀ
Alder adduct formation without significantly changing the kinetics
of dynamic exchange.
1
Figure 4. Partial H NMR spectra recorded at t = 0 h (top), t = 93 h
middle), and t = 400 h (bottom), demonstrating furan/2-methylfuran
(
exchange and equilibration of adducts 9 and 12.
Figure 2 summarizes the key results and predictions of the
computational investigations reported herein. Computed reac-
tion free energies (ΔG) and transition state free energies (ΔG )
q
further distinguished as being red, while exo isomers are blue.
Several notable trends can be ascertained from the plotted
reaction and transition state free energies: (i) in all cases the
exo isomer is the thermodynamic product, as indicated by a more
negative ΔG; (ii) in all cases except aldehyde substitution the
for DielsꢀAlder cycloadditions for each furan/maleimide com-
bination solvated in acetonitrile are plotted along the x and y axes,
respectively. Endo isomers of DielsꢀAlder products are indi-
cated by diamonds, while exo isomers of DielsꢀAlder products
are represented as squares. Products of the parent reaction
between furan and maleimide are shown in green. DielsꢀAlder
adducts involving substituted maleimide (N-methyl, N-vinyl, or
N-phenyl) are shown in gray. In all other cases endo isomers are
q
endo isomer is the kinetic product, as indicated by a lower ΔG
value; (iii) reactions between maleimide and 3-substituted furans
invariably lead to both more stable DielsꢀAlder adducts and
lower transition state barriers than the same reaction with the
7
999
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
corresponding 2-substituted furan; (iv) there is little variability
It is notable that the exchange of 2-methylfuran 3a with furan 1
occurs faster than exchange of N-methylmaleimide 6 with
maleimide 5. This experimental observation is in line with the
computationally predicted relative ratios of free energy barriers
for these reactions. Specifically, the barriers to cycloaddition of
2-methylfuran 3a with maleimide 5 are predicted to be 1.1ꢀ1.7
kcal/mol lower than the barriers to cycloaddition of unsubsti-
tuted furan with maleimide (5) or unsubstituted furan with N-
methylmaleimide (6). Transition state free energies of the last
two cycloaddition reactions—unsubstituted furan 1 reacting
with either maleimide 5 or N-methylmaleimide 6—are identical
within error ((0.5 kcal/mol). The observation that maleimide
exchange (Figure 3) is observed to require a longer time to reach
equilibrium than furan exchange (Figure 4) supports the com-
putationally predicted trend in free energy cycloaddition barriers.
These two representative examples of exchange processes
demonstrate the utility and practicality of dynamic exchange
involving substituted derivatives of furan dienes and maleimide
dienophiles. They lend support to the computational predictions
in this study and serve to further highlight the influences of furan
and maleimide substitution on the dynamic reversibility of their
DielsꢀAlder reactivity.
(
e1.0 kcal/mol) in the DielsꢀAlder reactivity of furan with
N-alkyl, N-vinyl, or N-phenyl maleimide; (v) the DielsꢀAlder
reactivity of maleimide with substituted furans varies considerably,
depending on the electronic character of the furan substituent and
whether the substituent is attached at the 2- or 3-position of furan.
1
H NMR Spectroscopy. With the above computational pre-
dictions at hand, it became of interest to experimentally study the
dynamic reversibility of furanꢀmaleimide DielsꢀAlder reactivity
through exhange reactions. Two dynamic exchange reactions
were investigated spectroscopically: (i) exchange of the malei-
mide moiety of adduct 9 with N-methylmaleimide (Scheme 3a)
and (ii) exchange of the furan moiety of adduct 9 with 2-methyl-
furan (Scheme 3b). For the dynamic exchange outlined in
Scheme 3a, computations predict an equilibrium ratio of 22:78
between adducts 9 and 16, given the relative stabilities of their
52
exo and endo adducts computed in acetonitrile at 348 K. A 0.25
M solution of furanꢀmaleimide adduct 9 and N-methylmalei-
mide 6 in CD CN was prepared in a screw-top NMR tube and
3
heated at 75 °C. Spectra were periodically recorded to monitor
the extent of maleimide exchange. Figure 3 shows a time
1
sequence of H NMR spectra demonstrating the dynamic
covalent exchange of maleimide with N-methylmaleimide. At
Conclusions. The importance of dynamic covalent chemistry
to materials organic synthesis has been well established and
continues to generate increasing interest. The application of
reversible DielsꢀAlder reactions in the dynamic covalent synth-
esis of organic materials has only recently been explored. The
results presented herein provide valuable insight into the elec-
tronic and regiochemical effects of furan and maleimide sub-
stitution on their applicability in dynamic covalent applications.
Care must be taken when selecting which diene and dienophile
combinations are used in dynamic covalent chemical applica-
tions, as differences in their electronic characteristics and sub-
stitution of the furan dienes at the 2- or 3-position greatly effect
their reaction and transition state energetics. Even with the
relatively limited series of monosubstituted derivatives studied
herein, it can be shown that furanꢀmaleimide reactivity can be
tuned across a considerable range of reactivity: varying from
largely exergonic and therefore irreversible reactions to mildly
exergonic reactions with favorable exchange kinetics to unfavorable
endergonic reactions. Selection of appropriate subsituents can help
optimize materials properties, depending on whether increased
reversibility, decreased reversibility, or particular exchange reactions
are desired for a particular application. Given the increasing use of
these systems in mendable macromolecules, such differences can
have, and are already being shown to have, pronounced implications
on material properties. Further investigations of a wider range of
substituents, disubstituted compounds, and additional dynamic
exchange competition studies are underway.
0
.25 M and 75 °C the exchange requires approximately 320 h
(see the Supporting Information). Heating a more concentrated
sample (2.0 M) of the same 1:1 mixture at 100 °C allows the
same equilibrium to be reached overnight as opposed to requir-
ing 8 days. As previously stated, dynamic covalent DielsꢀAlder
reactions are self-contained and only depend on temperature,
concentration, and solvent. Spectroscopically the ratio of furanꢀ
maleimide adduct 9 and furanꢀN-methylmaleimide adduct 16
is 36:64, as determined by integration of well-isolated signals
corresponding to hydrogen atoms H and H (Figure 3). This
experimentally measured ratio is in good agreement with the
computational prediction that N-methylmaleimide forms a
more stable adduct with furan than maleimide does; however,
experimental results do suggest computations may overestimate
the stability of adduct 16 relative to 9. Given the exponential
relationship between percentcontribution and relative free energy,
a small error in computed free energies can have a significant effect
on the predicted product ratios. Even so, the computationally
predicted ratio agrees quite well with the experimental adduct ratio
and, most importantly, computations and experiment agree that
adduct 16 is more stable than adduct 9.
3
6
Scheme 3b outlines the exchange of the unsubstituted furan
moiety of adduct 9 with 2-methylfuran 3a. Computations in
Table 3 predict the exo and endo isomers of 2-methylfuranꢀ
maleimide adduct 12 to be 0.4 and 1.3 kcal/mol more stable,
respectively, than the exo and endo isomers of unsubstituted
furanꢀmaleimide adduct 5. The ratio of adducts 5 and 12 at
equilibrium is therefore predicted to be 35:65. A 0.25 M CD CN
3
’
COMPUTATIONAL AND EXPERIMENTAL METHODS
solution containing equimolar amounts of furanꢀmaleimide
adduct 9 and 2-methylfuran 3a was sealed in a screw-cap NMR
Computational Details. All calculations were performed with the
7
tube and heated at 75 °C. Periodic monitoring of the H proton
53
0
Gaussian09 suite of programs. Prior to geometry optimization,
dihedral scans were performed at a low level of theory (HF/3-21G) to
approximate the global energy minimum conformation of molecular
species containing easily rotating torsion angles. Optimizations of
ground state geometries for global minimum conformations were then
8
8
signal of furanꢀmaleimide adduct 9 and the H and H proton
1
signals of 2-methylfuranꢀmaleimide adduct 12 by H NMR
spectroscopy indicated equilibrium was reached after 260 h (see
the Supporting Information). Again, a more concentrated sample
54
can be equilibrated overnight at 100 °C. Integration of protons
H and H
carried out to full convergence at the MP2/6-311+G(d,p) and CBS-
0
7
8+8
55
in Scheme 3b indicates an equilibrium ration of
QB3 levels of theory. Single-point electronic energies were computed
5
0:50 between adducts 12 and 9. The spectroscopic ratio is in
at the MP2/6-311+G(d,p) level using MP2 optimized geometries.
Vibrational analyses were carried out at the CBS-QB3 level, and the
general agreement with the computationally predicted ratio.
8
000
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
resulting frequencies were used for thermochemical calculations. Similar
’ REFERENCES
computational methodologies have been shown to be reliable for Dielsꢀ
(
(
1) Huc, I.; Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2106–2110.
2) Ramstrom, O.; Lehn, J.-M. Nat. Rev. Drug Discovery 2002,
49
Alder reactions involving furan and maleimide. Transition state searches
56
were performed using the QST2 method at the B3LYP/6-311G(d,p)
level and then refined by Berny optimization at the MP2/6-31+G(d)
and CBS-QB3 levels. Transition states were distinguished by having a
single imaginary vibrational frequency corresponding to the vibrational
mode that connects DielsꢀAlder adducts to their separated diene and
dienophile components. Gas-phase single-point electronic energies for
transition states were calculated at the MP2/6-31+G(d) level, with
vibrational analysis being carried out at the CBS-QB3 level for zero-
point and thermal corrections. Solvation free energies were calculated
1, 26–36.
(3) Lehn, J.-M.; Eliseev, A. V. Science 2001, 291, 2331–2332.
(4) Rowan, S. J.; Santrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.;
Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898–952.
(5) Corbett, P. T.; Leclaire, J.; Vial, L.; West, K. R.; Wietor, J.-L.;
Sanders, J. K. M.; Otto, S. Chem. Rev. 2006, 106, 3652–3711.
(
(
(
6) Lehn, J.-M. Angew. Chem., Int. Ed. 1990, 29, 1304–1319.
7) Lindsey, J. S. New J. Chem. 1991, 15, 153–180.
8) Philp, D.; Stoddart, J. F. Angew. Chem., Int. Ed. 1996, 35,
1
155–1196.
9) Leininger, S.; Olenyuk, B.; Stang, P. J. Chem. Rev. 2000, 100,
53–908.
57
utilizing the PCM reaction field model in acetonitrile (ε = 35.688).
Chemicals. Chemicals were obtained from commercial sources and
used as purchased. DielsꢀAlder adducts of unsubstituted as well as
derivatized furan and maleimide compounds were prepared according to
(
8
(
(
10) Grzybowski, B.; Whitesides, G. W. Science 2002, 295, 2418–2421.
11) See, for example: (a) Brady, P. A.; Bonar-Law, R. P.; Rowan,
3
8
literature procedures.
S. J.; Suckling, C. J.; Sanders, J. K. M. Chem. Commun. 1996, 319–320.
(b) Melson, G. A.; Busch, D. H. J. Am. Chem. Soc. 1964, 86, 213–217.
(c) Otto, S.; Furlan, R. L. E.; Sanders, J. K. M. Science 2002,
297, 590–593. (d) Fuchs, B.; Nelson, A.; Star, A.; Stoddart, J. F.; Vidal,
S. B. Angew. Chem., Int. Ed. 2003, 42, 4220–4224.
1
H NMR Spectroscopic Studies. Dynamic exchange reactions
1
were carried out directly in screw-cap NMR tubes. H NMR spectra
were obtained using a 300 MHz NMR spectrometer, and chemical shifts
are reported as parts per million (ppm) downfield from the signal of
(
(
12) Hubin, T. J.; Busch, D. H. Coord. Chem. Rev. 2000, 200, 5–52.
13) Albrecht, M. J. Inclusion Phenom. Macrocycl. Chem. 2000,
Me
4
Si. Equimolar amounts of an unsubstituted furanꢀmaleimide
DielsꢀAlder adduct and either N-methylmaleimide or 2-methylfuran
3
3
6, 127–151.
14) Wang, D. Y.; Mohwald, H. J. Mater. Chem. 2004, 14, 459–468.
were dissolved in 1.0 mL of CD CN in a screw-cap NMR tube to give a
3
(
0
.25 M solution. The top of the NMR tube was then wrapped tightly in
1
(15) Meyer, C. D.; Joiner, C. S.; Stoddart, J. F. Chem. Soc. Rev. 2007,
6, 1705–1723.
Teflon tape. An initial H NMR spectrum was obtained at 298 K
immediately after sample preparation and taken as time = 0 h. The NMR
tube was then suspended in a stirred mineral oil bath kept at a constant
(
16) Glink, P. T.; Oliva, A. I.; Stoddart, J. F.; White, A. J. P.; Williams,
D. J. Angew. Chem., Int. Ed. 2001, 40, 1870–1875.
17) Leigh, D. A.; Lusby, P. J.; Teat, S. J.; Wilson, A. J.; Wong, J. K. Y.
Angew. Chem., Int. Ed. 2001, 40, 1538–1543.
18) (a) Williams, A. R.; Northrop, B. H.; Chang, T.; Stoddart, J. F.;
White, A. J. P.; Williams, D. J. Angew. Chem., Int. Ed. 2006, 45, 6665–
669. (b) Northrop, B. H.; Aric ꢀo , F.; Tangchaivang, N.; Badji ꢀc , J. D.;
3
48 K, except when the sample was periodically removed from the oil
(
1
bath and cooled to room temperature and an H NMR spectrum was
obtained. Such monitoring of the dynamic exchange process initially
consisted of acquiring spectra at 1 h intervals, gradually ramping up to
longer time intervals between acquisitions. The integration of diagnostic
proton signals corresponding to different dynamic exchange products
was used to evaluate the extent of exchange as a function of time.
(
6
Stoddart, J. F. Org. Lett. 2006, 8, 3899–3902.
(19) Chichak, K. S.; Cantrill, S. J.; Pease, A. R.; Chiu, S.-H.; Cave,
G. W. V.; Atwood, J. L.; Stoddart, J. F. Science 2004, 304, 1308–1312.
(
20) Pentecost, C. D.; Chichak, K. S.; Peters, A. J.; Cave, G. W. V.;
Cantrill, S. J.; Stoddart, J. F. Angew. Chem., Int. Ed. 2007, 46, 218–222.
21) (a) C ^o t ꢀe , A. P.; Benin, A. L.; Ockwig, N. W.; O’Keeffe, M.;
’
ASSOCIATED CONTENT
(
Matzger, A. J.; Yaghi, O. M. Science 2005, 310, 1166–1170. (b) El-Kaderi,
H. M.; Hunt, J. R.; Mendoza-Cortes, J. L.; C ^o t ꢀe , A. P.; Taylor, R. E.;
O’Keeffee, M.; Yaghi, O. M. Science 2007, 316, 268–272.
S
Supporting Information. Figures, tables, and text giving
b
1
time versus mole fraction plots for H NMR dynamic exchange
experiments, comparison of H NMR spectra from dilute (0.25 M,
1
(22) (a) Kuhn, P.; Antonietti, M.; Thomas, A. Angew. Chem., Int. Ed.
4
00 h equilibration) and concentrated (2.0 M, 24 h equilibration)
2
008, 47, 3450–3453. (b) Thomas, A. Angew. Chem., Int. Ed. 2010,
dynamic exchange reactions, NMR spectra of adducts 14 and 15,
Cartesian coordinates of all stationary points reported in this paper
and their absolute energies in hartrees, and the full author list for
ref 53. This material is available free of charge via the Internet at
http://pubs.acs.org.
4
9, 8328–8344 and references therein.
(23) (a) Tilford, R. W.; Gemmill, W. R.; zur Loye, H. C.; Lavigne, J. J.
Chem. Mater. 2006, 18, 5296–5301. (b) Tilford, R. W.; Mugavero, S. J.;
Pellechia, P. J.; Lavigne, J. J. Adv. Mater. 2008, 20, 2741–2746.
(24) (a) Spitler, E.; Dichtel, W. R. Nature Chem. 2010, 2, 672–677.
(
b) Colson, J. W.; Woll, A. R.; Mukherjee, A.; Levendorf, M. P.; Spitler,
E. L.; Shields, V. B.; Spencer, M. G.; Park, J.; Dichtel, W. R. Science 2011,
32, 228–231. (c) Spitler, E. K.; Giovino, M. R.; White, S. L.; Dichtel,
W. R. Chem. Sci. 2011, 2, 1588–1593.
25) (a) Christinat, N.; Scopelliti, R.; Severin, K. J. Org. Chem. 2007,
2, 2192–2200. (b) Christinat, N.; Scopelliti, R.; Severin, K. Angew.
’
AUTHOR INFORMATION
3
Corresponding Author
(
*E-mail: bnorthrop@wesleyan.edu.
7
Chem., Int. Ed. 2008, 47, 1848–1852. (c) Christinat, N.; Scopelliti, R.;
Severin, K. Chem. Commun. 2008, 3660–3662.
(
26) (a) Nitschke, J. R. Acc. Chem. Res. 2007, 40, 103–122.
b) Hutin, M.; Bernardinelli, G.; Nitschke, J. R. Chem. Eur. J. 2008,
4, 4585–4593. (c) Campbell, V. E.; de Hatten, X.; Delsuc, N.;
Kauffmann, B.; Huc, I.; Nitschke, J. R. Nature Chem. 2010, 2, 684–687.
27) Sauer, J.; Sustmann, R. Angew. Chem., Int. Ed., Engl. 1980, 19,
779–807.
(28) (a) Chen, X.; Dam, M. A.; Ono, K.; Mal, A.; Shen, H.; Nutt,
S. R.; Sheran, K.; Wudl, F. Science 2002, 295, 1698–1792. (b) Chem, X.;
’
ACKNOWLEDGMENT
(
1
We are grateful to the ACS PRF for financial support through a
research grant to B.H.N. and to the ACS Physical Division
Workshop for Undergraduate Students for a travel grant to R.C.B.
We thank Wesleyan University for computer time supported by
the NSF under Grant No. CNS-0619508. We thank Prof. George A.
Petersson for helpful discussions.
(
8
001
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002

The Journal of Organic Chemistry
ARTICLE
Wudl, F.; Mal, A. K.; Shen, H.; Nutt, S. R. Macromolecules 2003, 36,
(57) (a) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981,
55, 117–129. (b) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem.
Phys. Lett. 1996, 255, 327–335.
1
802–1807.
29) (a) McElhanon, J. R.; Wheeler, D. R. Org. Lett. 2001,
, 2681–2683. (b) Szalai, M. L.; McGrath, D. V.; Wheeled, D. R.; Zifer,
(
3
T.; McElhanon, J. R. Macromolecules 2007, 40, 818–823. (c) Polasket,
N. W.; McGrath, D. V.; McElhanon, J. R. Macromolecules 2011, 44,
3
203–3210.
30) Merve Kose, M.; Yesilgad, G.; Sanyal, A. Org. Lett. 2008, 10,
353–2356.
31) Vieyres, A.; Lam, T.; Gillet, R.; Franc, G.; Castonguay, A.;
Kakkar, A. Chem. Commun. 2010, 1875–1877.
32) McElhanon, J. R.; Zifer, T.; Kline, S. R.; Wheeler, D. R.; Loy,
(
2
(
(
D. A.; Jamison, G. M.; Long, T. M.; Rahimian, K.; Simmons, B. A.
Langmuir 2005, 21, 3259–3266.
(
33) Tian, Q.; Yuan, Y. C.; Rong, M. Z.; Zhang, M. Q. J. Mater. Chem.
009, 19, 1289–1296.
34) Zhang, Y.; Broekhuis, A. A.; Picchioni, F. Macromolecules 2009,
2, 1906–1912.
35) Adzima, B.; Aguirre, H. A.; Kloxin, C. J.; Scott, T. F.; Bowman,
C. N. Macromolecules 2008, 41, 9112–9117.
36) Reutenauer, P.; Buhler, E.; Boul, P. J.; Candau, S. J.; Lehn, J.-M.
Chem. Eur. J. 2009, 15, 1893–1900.
37) (a) Liu, X.; Zhu, M.; Chen, S.; Yuan, M.; Guo, Y.; Song, Y.; Liu,
2
(
4
(
(
(
H.; Li, Y. Langmuir 2008, 24, 11967–11974. (b) Liu, X.; Liu, H.; Zhou,
W.; Zheng, H.; Yin, X.; Li, Y.; Guo, Y.; Zhu, M.; Ouyang, C.; Zhu, D.;
Xia, A. Langmuir 2010, 26, 3179–3185.
(
38) Zhu, J.; Kell, A. J.; Workentin, M. S. Org. Lett. 2006, 8,
993–4996.
39) Bahman, A.; Bakhtiari, S.; Hsiao, D.; Jin, G.; Gates, B. D.;
Branda, N. Angew. Chem., Int. Ed. 2009, 48, 4166–4169.
40) Lemieux, V.; Gauthier, S.; Branda, N. R. Angew. Chem., Int. Ed.
006, 45, 6820–6824.
4
(
(
2
(
(
41) Boul, P. J.; Reutenauer, P.; Lehn, J.-M. Org. Lett. 2005, 7, 15–18.
42) Kotha, S.; Banerjee, S.; Patil, M. P.; Sunoj, R. B. Org. Biomol.
Chem. 2006, 4, 1854–1856.
(
(
43) Jones, G. O.; Houk, K. N. J. Org. Chem. 2008, 73, 1333–1342.
44) Reutenauer, P.; Boul, P. J.; Lehn, J.-M. Eur. J. Org. Chem. 2009,
1
691–1697.
45) Pieniazek, S. N.; Houk, K. N. Angew. Chem., Int. Ed. 2006,
5, 1442–1445.
(
4
(46) Kwart, H.; Burchuk, I. J. Am. Chem. Soc. 1952, 74, 3094–3097.
(47) Berson, J.; Swidler, R. J. Am. Chem. Soc. 1954, 76, 4060–4069.
(48) Goh, Y. W.; Pool, B. R.; White, J. M. J. Org. Chem. 2008, 73,
1
51–156.
49) Rulísek, L.; Sebek, P.; Havlas, Z.; Hrabal, R.; Capek, P.; Svatos,
A. J. Org. Chem. 2005, 70, 6295–6302.
50) Tobia, D.; Harrison, R.; Phillips, B.; White, T. L.; DiMare, M.;
Rickborn, B. J. Org. Chem. 1993, 58, 6701–6706.
51) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger,
M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445–11459.
52) This ratio is based upon a Boltzmann distribution (at 348 K)
(
(
(
(
taking into account the relative free energies calculated for the four
products at equilibrium: 16-exo (ꢀ4.5 kcal/mol), 9-exo (ꢀ3.6 kcal/
mol), 16-endo (ꢀ1.5 kcal/mol), and 9-endo (ꢀ1.1 kcal/mol). Contribu-
tions of both exo and endo isomers of each adduct were combined to
arrive at the total 22:78 ratio for 9:16.
(
53) Frisch, M. J. et al. Gaussian 09, Revision A.1 (see the Supporting
Information for the full reference).
54) (a) Frisch, M. J.; Head-Gordon, M.; Pople, J. A. Chem. Phys.
(
Lett. 1990, 166, 275–280. (b) Frisch, M. J.; Head-Gordon, M.; Pople,
J. A. Chem. Phys. Lett. 1990, 166, 281–289.
(55) (a) Nyden, M. R.; Petersson, G. A. J. Chem. Phys. 1981,
75, 1843–1862. (b) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys.
1
991, 94, 6081–6090. (c) Petersson, G. A.; Tensfeldt, T.; Montgomery,
J. A. J. Chem. Phys. 1991, 94, 6091–6101. (d) Montgomery, J. A.;
Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1994, 101, 5900–5909.
(56) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput.
Chem. 1996, 17, 49–56.
8
002
dx.doi.org/10.1021/jo201606z |J. Org. Chem. 2011, 76, 7994–8002