Metal-free, noncovalent catalysis of Diels - Alder reactions by Neutral hydrogen bond donors in organic solvents and in water

Article abstract of DOI:10.1002/chem.200390042

We examined the catalytic activity of substituted thioureas in a series of Diels - Alder reactions and 1,3-dipolar cycloadditions. The kinetic data reveal that the observed accelerations in the relative rates are more dependent on the thiourea substituents than on the reactants or solvent. Although the catalytic effectiveness is the strongest in noncoordinating, nonpolar solvents, such as cyclohexane, it is also present in highly coordinating polar solvents, such as water. In 1,3-dipolar cycloadditions, the thiourea catalysts demonstrate only very moderate selectivity for reactions with inverse electron demand. Our experiments emphasize that both hydrophobic and polar interactions can co-exist, making these catalysts active, even in highly coordinating solvents. This class of catalysts increases the reaction rates and endo-selectivities of Diels - Alder reactions, in a similar manner to weak Lewis acids, without concomitant product inhibition.

Full text of DOI:10.1002/chem.200390042

FULL PAPER
Metal-Free, Noncovalent Catalysis of Diels Alder Reactions by Neutral
Hydrogen Bond Donors in Organic Solvents and in Water
Alexander Wittkopp[a] and Peter R. Schreiner*[a, b]
Abstract: We examined the catalytic
activity of substituted thioureas in a
series of Diels Alder reactions and
1,3-dipolar cycloadditions. The kinetic
data reveal that the observed accelera-
tions in the relative rates are more
dependent on the thiourea substituents
than on the reactants or solvent. Al-
though the catalytic effectiveness is the
strongest in noncoordinating, nonpolar
solvents, such as cyclohexane, it is also
present in highly coordinating polar
solvents, such as water. In 1,3-dipolar
cycloadditions, the thiourea catalysts
demonstrate only very moderate selec-
tivity for reactions with inverse electron
demand. Our experiments emphasize
that both hydrophobic and polar inter-
actions can co-exist, makingthese cata-
lysts active, even in highly coordinating
solvents. This class of catalysts increases
the reaction rates and endo-selectivities
of Diels Alder reactions, in a similar
manner to weak Lewis acids, without
concomitant product inhibition.
Keywords: catalysis ¥ Diels Alder
reaction ¥ hydrogen bonds ¥ hydro-
phobicity
remains uncertain. We will demonstrate herein that catalytic
activity can be achieved quite simply by means of properly
designed hydrogen-bond donors that activate the unsaturated
carbonyl moiety of, for instance, Michael-type dienophiles.
Introduction
Although there are many natural products, for example
catharanthine 1 and tabersonine 2,[1, 2] that may formally
Lewis acids, such as AlCl3 and TiCl4 accelerate Diels
24]
Alder reactions dramatically,[21
makingthem progress at
reasonable rates, even at low temperatures, while reducingthe
27]
amount of side products.[25 The rate enhancements can be
readily understood in terms of FMO theory. Upon coordina-
tion to a lone pair located on the Lewis-basic center of the
dienophile, the catalytically active Lewis acid withdraws
electron density and lowers the LUMO energy of the entire
(conjugated) system. This improves the HOMODiene
derive from intramolecular [4‡2]-cycloadditions (there are
naturally occurringproteins which catalyze Diels Alder
reactions[3, 4]), there is no definitive proof that these reactions
actually take place in biosynthesis.[5] Notwithstanding, there
are many examples of catalytic Diels Alder reactions under
conditions reminiscent of biological systems, such as RNA-
31]
LUMODienophile interaction.[28
More modern approaches
that combine experiment and theory show that Lewis acids
strongly affect the geometries of the transition structures, that
is, makingthem more asynchronous; these reactions are
nevertheless considered to be concerted.[32, 33] A comparison
of computed and experimental H/D and 13C/12C kinetic
isotope effects is a particularly effective tool to probe the
nature of the experimental transition structure of the pa-
14]
based mixtures of metals,[6, 7] catalytic antibodies,[8 encap-
[15 17]
sulatingpolysaccharoids,
and selfassemblingmolecular
capsules.[18 20] While the catalytic activity is usually ascribed to
hydrogen bonding, hydrophobicity, and other effects or
combinations thereof, the exact type of interaction often
39]
rent[34] and catalyzed Diels Alder reactions.[35
While the effects of hydrogen bonding in protic solvents or
active sites may be rationalized similarly, the bindingenergies
are expected to be much smaller. This is not undesirable
because strongbindingof the startingmaterials or products
does not necessarily lead to rate accelerations. Quite the
contrary, bindingof the product often leads to product
[a] Prof. Dr. P. R. Schreiner, A. Wittkopp
Institut f¸r Organische Chemie, Georg-August-Universit‰t Gˆttingen
Tammannstrasse 2, 37077 Gˆttingen (Germany)
E-mail: prs@org.chemie.uni-giessen.de
[b] Prof. Dr. P. R. Schreiner
inhibition that is often observed in simple Lewis acid catalyst
New address: Institute of Organic Chemistry, Justus-Liebig-University,
Heinrich-Buff-Ring58
44]
reactions (vide infra);[40] exceptions are known.[41
Hence,
an ideal bindingsituation consists of a variety of interactions
35392 Giessen (Germany)
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0407 $ 20.00+.50/0
407

P. R. Schreiner and A. Wittkopp
FULL PAPER
(hydrogen bonding plus hydrophobic effects[45, 46]) which all
are relatively small, but complementary if the overall fit to the
transition state is to be maximized relative to reactants and
products.[8]
Besides the classic monodentate Lewis acids there are also
polydentate Lewis acids which often doubly coordinate
carbonyl functionalities; one of many examples is the complex
of bidentate 1,2-phenylenedimercury derivatives with the
carbonyl group of dimethylformamide.[47] Not surprisingly, a
similar coordination pattern is found for the complexation of
carbonyl compounds with small hydrogen-bond donors, such
as water itself. Jorgensen×s theoretical rationalization of the
experimentally observed acceleration of the Claisen rear-
rangement in water also suggests that carbonyl groups accept
two H bonds (3).[48, 49] Experimental confirmation of this
theoretical prediction is available through the X-ray struc-
tures of the adduct of the biphenylenediol (4) with 2,6-
dimethylpyran-4-one[50, 51] and the complexes of a m-nitro-
diaryl urea with several Lewis bases.[52]
solvents,[59] very often they are not even needed.[60] Again, the
primary factors for the frequently remarkable changes in
increased reaction rates, stereochemistry, and chemoselectiv-
ities in these hydrogen-bonding environments are assigned to
64]
hydrogen bonding and hydrophobic effects.[46, 61
We demonstrate in the followingthat bidentate hydrogen-
bond donors, like certain substituted thioureas, catalyze
Diels Alder reactions with a catalytic loadingof 1 mol%.
As thiourea derivatives are generally more soluble in a wide
range of solvents, long alkyl chains as in 5 are dispensable.[55]
The hydrogen-bond donor ability of thiourea derivatives to
carbonyl groups is expected from the enhanced differences in
acidities {pKa thiourea ˆ 21.0; pKa urea ˆ 26.9}.[65] Further-
more, the lower electronegativity of sulfur makes self-
À
association (interaction of the N H group of one molecule
with the carbonyl or thiocarbonyl group of another) less
favorable. As these model catalysts are easy to synthesize, we
investigated the catalytic effectiveness of a series of sym-
metrically substituted thioureas 7a n (Scheme 1). Several
1,3-disubstituted alkyl, cycloalkyl, and phenyl thioureas were
tested to screen their catalytic abilities. Since different
substituents with possibly opposingeffects would unnecessa-
rily cloud our analysis, we only utilized symmetrical thioureas.
We also prepared two alkyl derivatives with n-octyl (7a) and
cyclohexyl groups (7b) for comparison.
The similarity between Lewis acids and hydrogen-bond
donors is also apparent from their ability to accelerate and
stereochemically alter organic transformations.[53] Hence,
some well-chosen bidentate hydrogen-bond donors increase
the reaction rates and change the stereochemical course of
some reactions. For instance, 40 mol% of 4 increases the
yields of Diels Alder reactions of a,b-unsaturated carbonyl
compounds with cyclopentadiene by a factor of 12.[54] An
equimolar amount of the sub-
To be systematic, we examined aniline derivatives with
electron-donatingand electron-withdrawinggroups in various
stituted diphenylurea 5 causes
up to fivefold rate enhance-
ment in Claisen rearrange-
ments.[55] The same reagent
S
S
S
N
H
N
H
N
H
N
H
7b
7e
7a
(20 mol%) increases the endo/
S
exo selectivity in allylations of
a-sulfinyl radicals by a factor of
1.5.[56] The catalytic activity of
the amidinium ion 6 is compa-
rable to that of mild Lewis acids
acceleratingsome Diels Alder
reactions by a factor of 1.7
450;[57] this is largely attributed
to an increased interaction of
S
S
N
H
N
H
N
H
N
H
N
H
N
H
7c
7f
7d
CF3
CF3
F3C
CF3
S
S
S
S
S
S
N
H
N
H
N
H
N
H
N
H
N
H
CF3
CF3
7h
7k
7n
7g
7j
F
F
F
F
S
S
À
the highly polarized N H
bonds in the cation.
N
H
N
H
N
H
N
H
N
H
N
H
F
F
In a similar manner to these
designed bidentate reagents,
protic solvents, such as water,
also accelerate pericyclic reac-
tions.[58] Although organic and
organometallic additives are
still active in protic and polar
7i
CF3
CF3
Cl
Cl
Cl
Cl
N
H
N
H
F3C
N
H
N
H
CF3
Cl
7m
N
H
N
H
Cl
Cl
Cl
7l
Scheme 1. The symmetrically substituted thioureas used in the present work.
408
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0408 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Metal-Free, Noncovalent Catalysis of Diels Alder Reactions
407 414
ringpositions. Our expectation was that noncoordinating
electron-withdrawing groups (e.g., CF3) in the meta positions
of the ringwould enhance the hydrogen bondingability of the
thiourea derivatives (7a n, 1 mol% in all experiments) and
without an additive in deuterated chloroform. Although
chloroform is an hydrogen-bond donor and is known to
accelerate Diels Alder reactions[68] (compared to reactions
in nonpolar, non-coordinatingsolvents, vide infra) we utilized
it in our reactions because it dissolves a wide range of organic
compounds. Furthermore, its hydrogen-bond donor abilities
are rather poor so that its rate-acceleratingeffects should
easily be overcome by those of the catalyst. We also carried
out reactions in cyclohexane and water to unveil solvent
effects and to separate them from the effectiveness of the
catalyst. With a 10-fold excess of cyclopentadiene, all
reactions were strictly pseudo-first-order and the relative rate
constants krel were determined by least-error square fits of the
kinetic data. While krel strongly depends on the choice of
catalyst, the reaction order is nearly constant over time
(Figure 1), indicating the absence of product inhibition.
Most thioureas under investigation show catalytic behavior
and increase the rate of product formation significantly, even
at a catalytic loadingof only 1 mol%. For instance, adding
1 mol% of catalyst 7g increases the reaction rate by a factor
À
N H bonds. Such electron-withdrawingsubstituents are the
trifluoromethyl group (monosubstituted in para, meta, and
ortho positions; 7 f h) and the fluoro group (also monosub-
stituted in para, meta, and ortho positions; 7i k). To
determine the effect of double substitutions on the catalyst
quality we also examined thioureas derived from the respec-
tive aniline derivatives (7l n). It is noteworthy that the
hydrogen-bonding properties of ureas play an important role
in their functions as herbicides, solubilizers, inclusion com-
pounds, nonlinear optical materials,[66] and as HIV-protease
inhibitors.[67
Results and Discussion
Catalyst evaluation: To screen catalyst efficiencies, we carried
out a series of Diels Alder reactions of cyclopentadiene with
several a,b-unsaturated carbonyl compounds catalyzed by
Figure 1. Kinetic data of the Diels Alder reaction of methyl vinyl ketone and cyclopentadiene catalyzed by the thiourea derivatives 7. The curves result
from fits of least-error square fits.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0409 $ 20.00+.50/0
409

P. R. Schreiner and A. Wittkopp
FULL PAPER
Table 1. Relative rate constants krel of the reaction of cyclopentadiene (9, 10-fold excess) with the dienophiles 8a e.
Entry
1
Dienophile
7
a
b
c
d
e
f
g
h
i
j
k
l
m
n
1.0
1.2
1.4
1.4
1.3
2.4
2.5
1.1
2.9
1.4
1.2
4.8
3.3
1.0
2
3
4
5
1.2
0.9
1.0
1.1
1.0
1.1
1.3
1.2
1.1
1.1
1.3
1.5
1.3
1.2
1.5
1.4
1.1
1.2
1.2
1.2
1.2
1.3
1.6
1.9
2.4
3.1
2.2
5.9
1.3
1.2
1.3
1.1
1.7
3.1
1.9
5.2
1.6
3.3
2.0
4.1
1.1
1.0
1.4
1.7
4.3
5.1
5.3
8.2
2.7
2.7
2.1
3.5
1.1
1.0
1.1
4.0
of ꢀ6 (Table 1, entry 5). Furthermore, product inhibition
seems to be very minor because the activity is still present
even after 80% conversion. The highly dynamic interactions
between the catalysts, startingmaterials, and products appa-
rently make the catalysts less susceptible to product inhib-
ition.
because of the electron-withdrawingrgoups (EWGs). The
À
small hydrogen-bond donor ability of these C H bonds leads
to internal interactions between the Lewis-basic sulfur and the
ortho hydrogen atoms which hinder the rotation of the phenyl
groups (Scheme 2).[70]
As expected, the choice of catalyst is crucial. While the
alkyl-substituted thioureas (7a and 7b) and 1,3-diphenylth-
iourea (7c) only cause minimal rate accelerations, the
acceleratingeffects of the fluorophenyl-substituted thiourea
(7j) and especially the trifluoro-substituted (7 f l) thioureas
are appreciable.
attractive S · · · H interaction
EWG
EWG
repulsive S · · · EWG interaction
EWG
EWG
H
H
S
S
Since the complexation between thioureas and carbonyl
compounds is modestly strong( ꢀ7 kcalmolÀ1 at room tem-
perature in dichloromethane)[53, 69] the complexation con-
stants are likely to be dominated by entropic effects that may
surpass the bindingexothermicities. This also implies that the
strength of the interaction depends on the rigidity of the
catalyst. Thus, the poor performance of 7a and 7b is
entropically unfavorable since the floppy alkyl substituents
of the free thiourea derivatives have to be ordered for proper
interaction of the catalyst with the dienophile. A similar
argument applies to the rotation of the aryl groups in 7c and
7d. In uncomplexed 7c or 7d, the rotational barrier is small
(accordingto B3LYP/6-31G* computations the barrier of 7c
is only 1.5 kcalmolÀ1, while the barrier of rotation in 7l is
3.4 kcalmolÀ1)[70] resultingin entropy loss upon bindingand
hence leading to a negligible catalytic effectiveness. For the
same reason, all ortho-substituted phenylthioureas 7e, h, k, n
are also less active. In the case of the thioureas derived from
meta- and para-substituted anilines, 7 f, g, i, j, l, m, the hydro-
gen atoms in the ortho position are more positively polarized
H
H
H
H
Scheme 2. The interactions influencingthe rotation of the phenyl groups.
In summary, thiourea derivatives with rigid electron-with-
drawingaromatic substituents are the most effective H-
bondingcatalysts for Diels Alder reactions considered in the
present study. The electron-withdrawingsubstituents in meta-
or para- positions aid in reducingthe flexibility of the catalyst,
thus minimizingthe entropic penalty upon complexation. The
acceleratingeffect can even be magnified by disubstituted
electron-poor phenyl groups, making 7l and 7m some of the
more efficient catalysts in this series.
These new types of catalysts are also active for other Diels
Alder reactions with a,b-unsaturated carbonyl compounds
(Table 1). The relative effectiveness does not depend on the
reaction, that is, the relative catalyst efficiencies are the same
in different reactions: 7g, 7l, and 7m are consistently the best
catalysts as they increase the relative rates by factors of 3 8;
410
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0410 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Metal-Free, Noncovalent Catalysis of Diels Alder Reactions
407 414
Table 2. Yields and selectivities (after 20 h) of Diels Alder reactions between cyclopentadiene (9, 10-fold
excess, 408C) and a,b-unsaturated aldehydes with and without 20 mol% 7g. Values in parentheses are from
references [74, 75].
dienophile 8e may be consid-
ered to be special case as it
could form a doubly hydrogen-
bonded complex.[71]
In a separate series of experi-
ments we examined the
changes of the exo/endo selec-
tivities in reactions of two a,b-
unsaturated compounds with
Entry Dienophile Yield without cat.[%] Yield with cat.[%] endo:exo without cat. endo:exo with cat.
cyclopentadiene
(Table 2).
Metallic Lewis acids generally
increase the selectivity of
Diels Alder reactions mostly
in favor of the endo product.[72]
For instance, AlCl3 ¥ OEt2 im-
proves the endo selectivity of
the reaction of cyclopentadiene
and methyl acrylate from 82%
to 98%.[73] Since the analysis of
1
2
16
31
24
66
65:35 (62:38)
18:82 (17:83)
77:23
8:92
Table 3. Relative rate constants krel of the reaction of cyclopentadiene (9) with 8a.
1
the endo/exo ratio by H NMR
techniques is simplified when
aldehydes instead of ketones
are used, we chose crotonalde-
hyde (8 f; endo selectivity, en-
try 1, Table 2) and methacrolein
(8g; exo selectivity, entry 2, Ta-
ble 2) as the dienophiles.
In this set of reactions, thio-
urea 7g also reveals its similar-
ity with mild Lewis acids.[53] In
7
a
b
c
d
e
f
g
h
i
j
k
l
m
n
1.1
1.8
1.9
2.0
1.8
3.7
3.9
1.3
3.3
2.0
1.4
8.8
5.1
1.5
the case of crotonaldehyde, the yield obtained after 20 h is
increased by a factor of 1.5 and the endo selectivity increases
from 65% to 77%. For methacrolein we found a 2.1-fold yield
increase (also after 20 h) of the uncatalyzed reaction while the
exo selectivity changed from 82% to 92%.
Effect of solvent: As chloroform itself is a weak hydrogen-
bond donor capable of accelerating[4 ‡2] cycloadditions to
some degree, we carried out a selected set of Diels Alder
reactions of methyl vinyl ketone and cyclopentadiene with
7a n in cyclohexane. This solvent is expected to have
negligible interactions with the solutes, allowing direct
observation of the rate enhancements by the respective
catalyst (Table 3). As a consequence, the catalysts show the
same order of activity while the increases in the reaction rates
are even higher (with krel up to 9, for 7l).
1
Figure 2. Product formation over time ([%], H NMR) in the reaction of
8a with 9 in three solvents without and with 1 mol% catalyst 7g. The
steeper slopes refer to the catalyzed reaction. The data are fitted on the
basis of a first-order reaction (see the Experimental Section).
While strongLewis acids often cannot be used in protic or
highly polar solvents because of hydrolysis or strong solvation
Table 4. Product yields in three solvents with and without catalyst 7g in the
reaction of 8a with 9.
that completely deactivates the catalysts (again, exceptions
83]
are known[71, 76, 77]),[78
and while only rather mild Lewis
Solvent
Catalyst
[mol%]
Yield after 1 h
[%][a]
acids with metal centers, such as [Yb(OTf)3], are active in
86]
such solvents,[83
the much lowered acceptor strength of 7
cyclohexane
chloroform
18
42
31
52
65
74
85
should make these catalysts less sensitive to competitive
aqueous solvation. This was confirmed by the reaction of
cyclopentadiene (9) and methyl vinyl ketone (8a) in cyclo-
hexane, chloroform, and water (Figure 2, Table 4). This
reaction is particularly suitable since it is easily monitored
by NMR spectroscopy (the product 1H NMR resonance
signals are well separated from those of the starting materials)
1
1
40
water (incl. 10 vol% tBuOH)
1
1
[a] Æ 1%, determined by H NMR spectroscopy.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0411 $ 20.00+.50/0
411

P. R. Schreiner and A. Wittkopp
FULL PAPER
[94]
and because of its reaction rate, which is fast enough to obtain
the data within a reasonable timeframe and slow enough to be
observable on the NMR time scale.
LUMOdipolarophile
.
Complexation of the most Lewis-basic
site, the nitrone oxygen, would lower the energy of the
HOMOnitrone and hence decelerate the reaction. On the other
hand, complexation of the nitrone in a 1,3-dipolar cyclo-
addition with inverse electron demand (e.g., between a
nitrone and a a,b-unsaturated ether where the LUMOnitrone
is interactingwith the HOMO dipolarophile) would reduce the
HOMO LUMO gap and thus accelerate the reaction. Hence,
if our assumption that hydrogen-donor complexation is
mainly responsible for the observed thiourea catalysis is
correct, 1,3-dipolar cycloadditions with inverse electron
demand should be accelerated. To test this hypothesis, we
examined the reactions of 11 with 8a and with isopropyl vinyl
ether (12).
Figure 2 demonstrates that in cyclohexane, chloroform, and
water (with 10% tert-butyl alcohol to solubilize all reac-
tants[87]) the reaction orders were constant and of pseudo-first
order. While catalyst 7g is active in all chosen solvents, it is
particularly strikingthat acceleration is also observed in
water, even in catalyst concentrations as low as 1 mol%!
Protic solvents, for example alcohols, generally accelerate
Diels Alder reactions[88] by H-bondingand by reducingthe
HOMO LUMO gap of the reactants. In water, however, not
only the polarity of the solvent but also the hydrophobicities
91]
of the solutes are important.[87
This rather complex
property is governed by the limited ability of water to dissolve
nonpolar molecules and is considered to be important in
enzyme substrate interactions,[10] the aggregation of amphi-
philic molecules into supramolecular structures (e.g., micelles
and vesicles),[92] molecular recognition phenomena, and sur-
face forces.[93] It causes nonpolar molecules or parts thereof to
agglomerate to small capsules in aqueous media.[63]
Metallic Lewis acids, on the other hand, are rather polar
and, if not hydrolyzed, are often highly solvated in aqueous
media leadingto a reduction of their catalytic effectiveness.
Hence, a possible explanation for our findings is that the
hydrophobic organic molecules are forced together in water
that interacts better with itself than with the solute. This effect
is cumulative to the TS-stabilizingpolar interactions of water
with the transition state. These conclusions compare well with
a very recent computational study of the solvent effects on the
Diels Alder reaction of butadiene with acrolein.[46]
While addition of an equimolar amount of any of our
thiourea derivatives does not have an acceleratingeffect on
the 1,3-DC with normal electron demand (8a, Table 5) or may
even be considered decelerating, the catalysts are very
modestly effective in the reaction with inverse electron
demand (12, Table 5). Again, thiourea 7l is one of the more
efficient additives; after 10 h, the yield is nearly doubled
compared to the uncatalyzed reaction.
Table 5. The yield ratio of the catalyzed and the uncatalyzed 1,3-dipolar
cycloadditions of above reactants; t ˆ 20 h, Tˆ 608C, 1 equiv catalyst,
10 equiv dipolarophile.
The authors found that the microsolvation effect of explicit
water molecules induces a polarization and hence energy
loweringof the transition structure. In the spirit of the
Jorgensen model (vide supra), our catalysts take the place of
the 2 3 required water molecules. The aforementioned
computational study also showed that this hydrogen-bonding
microsolvation accounts for approximately one-half of the
observed catalytic effect. The other half stems from increased
hydrophobic interactions which are maximized in water as the
solvent around both the reactants and the catalysts in our case.
The explicit hydrogen bonding is necessary to induce charge
polarization of the transition structure and to allow enforced
hydrophobic interactions. The endo/exo selectivity is equally
influenced by both hydrogen bonding and bulk phase
effects.[46] The present study, therefore, is a welcome con-
firmation of the computational analysis and emphasizes the
possibility that both effects, polarization and hydrophobicity,
can indeed be complementary rather than competitive.
Dipolarophile
7
f
g
h
i
j
k
l
8a
12
1.0
1.6
0.9
1.5
0.9
1.0
1.0
1.2
1.0
1.4
1.1
1.0
0.8
1.8
The observation that our catalysts do accelerate the 1,3-
dipolar cycloaddition with inverse but not the one with normal
electron demand, supports our proposition that the catalysts
operate by hydrogen-bonding to the most Lewis-basic groups
and reducing the HOMO LUMO gap in analogy to mild
Lewis acids.
À
Finally, we also probed catalysts with only one N H bond
donor (the N-monomethylated forms of 7g and 7l and found
them virtually ineffective. This is supported by the clamplike
bindingmotif (i.e., Jorgensen×s water model) found for these
[53]
types of catalysts when interactingwith carbonyl groups.
1,3-Dipolar cycloadditions with normal and inverse electron
demand: To verify the hypothesis that the interaction between
the thiourea and the carbonyl groups causes the observed
accelerations, we examined the catalytic activity of 7 f l in
1,3-dipolar cycloadditions of N-benzylideneaniline N-oxide
(nitrone 11) with dipolarophiles. In 1,3-dipolar cycloadditions
with normal electron demand (e.g., between a nitrone and
a,b-unsaturated carbonyl compounds), the most important
FMO interaction is the overlap of the HOMOnitrone with the
Conclusion
Simple, neutral hydrogen-bond donors, such as substituted
thioureas, are able to catalyze Diels Alder reactions and 1,3-
dipolar cycloadditions, increasingthe reaction rates and
stereoselectivities. The relative effectiveness of these catalysts
depends more on their substituents than on the reactants or
412
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0412 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Metal-Free, Noncovalent Catalysis of Diels Alder Reactions
407 414
[5] S. Laschat, Angew. Chem. 1996, 108, 313 315; Angew. Chem. Int. Ed.
Engl. 1996, 35, 289 291.
solvent. Generally, highly flexible hydrogen-bond donors
suffer from entropy loss upon complexation that cannot be
overcome by enthalpic effects. In contrast, more rigid
thiourea derivatives bind more favorably and, as a conse-
quence of their electron-poor nature, reduce the reaction
barriers. As there is a very fine balance includingsmall
enthalpic binding, product inhibition is not observed.
Counterintuitively, a highly coordinating solvent, such as
water, does not override the catalytic effect. While the
reactions are accelerated by polar interactions of water, even
without a catalyst, addition of only 1 mol% catalyst still leads
to detectable rate accelerations, that is, solvent and additive
do not necessarily compete, but may operate in a comple-
mentary fashion, as predicted computationally.[46]
[6] B. Seelig, S. Keiper, F. Stuhlmann, A. J‰schke, Angew. Chem. 2000,
112, 4764 4768; Angew. Chem. Int. Ed. 2000, 39, 4576 4579.
[7] T. M. Tarasow, S. L. Tarasow, B. E. Eaton, Nature 1997, 389, 54 57.
[8] J. G. Chen, Q. L. Deng, R. X. Wang, K. N. Houk, D. Hilvert,
Chembiochem 2000, 1, 255 261.
[9] J. Hasserodt, Synlett 1999, 12, 2007 2022.
[10] D. Hilvert, Annu. Rev. Biochem. 2000, 69, 751 793.
[11] D. Hilvert, K. W. Hill, K. D. Nared, M.-T. M. Auditor, J. Am. Chem.
Soc. 1989, 111, 9261 9262.
[12] R. A. Lerner, S. J. Benkovic, P. G. Schultz, Science 1991, 252, 659
667.
[13] J. A. Xu, Q. L. Deng, J. G. Chen, K. N. Houk, J. Bartek, D. Hilvert,
I. A. Wilson, Science 1999, 286, 2345 2348.
[14] A. Heine, E. A. Stura, J. T. Yli-Kauhaluoma, C. S. Gao, Q. L. Deng,
B. R. Beno, K. N. Houk, K. D. Janda, I. A. Wilson, Science 1998, 279,
1934 1940.
[15] N. K. Sangwan, H.-J. Schneider, J. Chem. Soc. Perkin Trans. 2 1989,
1223 1227.
[16] H. J. Schneider, N. K. Sangwan, Angew. Chem. 1987, 99, 924 925;
Angew. Chem. Int. Ed. Engl. 1987, 26, 896 897.
[17] H.-J. Schneider, N. K. Sangwan, J. Chem. Soc. Chem. Commun. 1986,
1787 1739.
Hence, we have identified a neutral system that relays its
ability to lower the activation energies of a subset of Diels
Alder reactions through specific hydrogen bonds.[53] Finally,
this allows the bold question as to whether the ™hunt∫ for a
™Diels Alder-ase∫ is still meaningful.[5]
[18] Y. L. Cho, D. M. Rudkevich, A. Shivanyuk, K. Rissanen, J. Re-
bek, Jr., Chem. Eur. J. 2000, 6, 3788 3796.
[19] J. Kang, G. Hilmersson, J. Santamaria, J. Rebek, Jr., J. Am. Chem.
Soc. 1998, 120, 3650 3656.
Experimental Section
Materials: All thiourea derivatives,[95] N-benzylidene-aniline N-oxide,[96, 97]
1,3-diphenyl-propenone,[98] 3-phenyl-1-pyridin-2-yl-propenone,[99] and
3-phenyl-1-pyridin-3-yl-propenone[100] were synthesized followingthe pro-
cedures reported in the literature. Methyl vinyl ketone (Fluka), isopropyl
vinyl ether (Acros), methacrolein (Acros), and crotonaldehyde (Acros)
were distilled immediately before use. Cyclopentadiene was prepared from
its dimer (Aldrich) by thermal retro-Diels Alder reaction immediately
before use. All solvents used were of the highest purity available.
Deuterated chloroform was stored over sodium bicarbonate.
[20] J. Kang, J. Santamaria, G. Hilmersson, J. Rebek, Jr., J. Am. Chem.
Soc. 1998, 120, 7389 7390.
[21] H. L. Goering, C. S. Chang, J. Org. Chem. 1975, 40, 2565 2565.
[22] P. V. Alston, R. M. Ottenbrite, J. Org. Chem. 1975, 40, 1111
1116.
[23] K. Williams, Y. F. L. Hsu, J. Am. Chem. Soc. 1970, 92, 7385 7389.
[24] L. F. Tietze, C. Ott, U. Frey, Liebigs Ann. 1996, 63 67.
[25] F. Fringuelli, O. Piermatti, F. Pizzo, L. Vaccaro, Eur. J. Org. Chem.
2001, 439 455.
[26] L. C. Dias, J. Braz. Chem. Soc. 1997, 8, 289 332.
[27] T. R. Kelly, S. K. Maity, P. Meghani, N. S. Chandrakumar, Tetrahe-
dron Lett. 1989, 30, 1357 1360.
[28] K. N. Houk, Top. Curr. Chem. 1979, 79, 1 40.
[29] K. N. Houk, Acc. Chem. Res. 1975, 8, 361 369.
[30] B. Depascualteresa, J. Gonzalez, A. Asensio, K. N. Houk, J. Am.
Chem. Soc. 1995, 117, 4347 4356.
[31] D. M. Birney, K. N. Houk, J. Am. Chem. Soc. 1990, 112, 4127 4133.
[32] K. N. Houk, J. Gonzalez, Y. Li, Acc. Chem. Res. 1995, 28, 81 90.
[33] Y. Li, K. N. Houk, J. Am. Chem. Soc. 1993, 115, 7478 7483.
[34] K. N. Houk, Y. Li, J. D. Evanseck, Angew. Chem. 1992, 104, 711 739;
Angew. Chem. Int. Ed. Engl. 1992, 31, 682 708.
[35] J. W. Storer, L. Raimondi, K. N. Houk, J. Am. Chem. Soc. 1994, 116,
9675 9683.
[36] K. N. Houk, Y. Li, J. Storer, L. Raimondi, B. Beno, J. Chem. Soc.
Farad. Trans. 1994, 90, 1599 1604.
[37] O. Wiest, K. N. Houk, K. A. Black, B. Thomas, J. Am. Chem. Soc.
1995, 117, 8594 8599.
Kinetics: All NMR measurements were recorded on a Bruker Aspect 300
NMR spectrometer. The Diels Alder reactions were carried out in a 5-mm
standard NMR tube containing0.5 mL of a solution of 0.1 m dienophile,
1.0 m 9, and the respective amount of catalyst (in the case of ketone
dienophiles 8a e 1 mol%, in the case of aldehyde dienophiles 8 f,g
20 mol%) in the chosen solvent. Since the solubility of 9 and the thioureas
in water is very low, 0.1 equiv tert-butyl alcohol were used as solubilizers for
all reactions in water (also in those without catalyst for proper comparison).
The Diels Alder reactions of 8a,b were carried out at 208C and were
recorded over 2 h. All other Diels Alder reactions were carried out at
408C and recorded over 40 h. Pseudo-first-order rate constants were
calculated with a fittingprogram. The 1,3-dipolar cycloaddition reactions
were also carried out in a 5-mm standard NMR tube containing0.5 mL of a
solution of 0.1m 11, 1.0m dipolarophile, and 0.1m catalyst in deuterated
chloroform at 608C and analyzed after 20 h. All reactions were repeated
and the results were averaged.
[38] D. A. Singleton, S. R. Merrigan, B. R. Beno, K. N. Houk, Tetrahe-
dron Lett. 1999, 40, 5817 5821.
Acknowledgements
[39] B. R. Beno, K. N. Houk, D. A. Singleton, J. Am. Chem. Soc. 1996,
118, 9984 9985.
[40] I. R. Hunt, A. Rauk, B. A. Keay, J. Org. Chem. 1996, 61, 751 757.
[41] H. Yamamoto, A. Yanagisawa, K. Ishihara, S. Saito, Pure Appl.
Chem. 1998, 70, 1507 1512.
This work was supported by the Deutsche Forschungsgemeinschaft
(Schr 597/3 1 and 3 2) and the Fonds der Chemischen Industrie. AW
thanks the Nieders‰chsische Graduiertenfˆrderungand the DAAD for
financial support.
[42] S. Kobayashi, Eur. J. Org. Chem. 1999, 15 27.
[43] M. Wills, J. Chem. Soc. Perkin Trans. 1 1998, 3101 3120.
[44] H. B. Kagan, O. Riant, Chem. Rev. 1992, 92, 1007 1019.
[45] T. Lazaridis, Acc. Chem. Res. 2001, 34, 931 937.
[46] S. Kong, J. D. Evanseck, J. Am. Chem. Soc. 2000, 122, 10418 10427.
[47] J. D. Wuest, Acc. Chem. Res. 1999, 32, 81 89.
[48] W. L. Jorgensen, J. F. Blake, D. C. Lim, D. L. Severance, J. Chem.
Soc. Farad. Trans. 1994, 90, 1727 1732.
[1] N. Kato, X. Wu, H. Nishikawa, K. Nakanishi, H. Takeshita, J. Chem.
Soc. Perkin Trans. 1 1994, 1047 1053.
[2] U. Pindur, G. H. Schneider, Chem. Soc. Rev. 1995, 23, 409 415.
[3] K. Auclair, A. Sutherland, J. Kennedy, D. J. Witter, J. P. Van den He-
ever, C. R. Hutchinson, J. C. Vederas, J. Am. Chem. Soc. 2000, 122,
11519 11520.
[49] D. L. Severance, W. L. Jorgensen, J. Am. Chem. Soc. 1992, 114,
10966 10968.
[4] K. Watanabe, T. Mie, A. Ichihara, H. Oikawa, M. Honma, J. Biol.
Chem. 2000, 275, 38393 38401.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0413 $ 20.00+.50/0
413

P. R. Schreiner and A. Wittkopp
FULL PAPER
[50] J. Hine, K. Ahn, J. Org. Chem. 1987, 52, 2083 2086.
[51] J. Hine, K. Ahn, J. C. Gallucci, S.-M. Linden, J. Am. Chem. Soc. 1984,
106, 7980 7981.
[52] M. C. Etter, S. M. Reutzel, J. Am. Chem. Soc. 1991, 113, 2586 2598.
[53] P. R. Schreiner, A. Wittkopp, Org. Lett. 2002, 4, 217 220.
[54] T. R. Kelly, P. Meghani, V. S. Ekkundi, Tetrahedron Lett. 1990, 31,
3381 3384.
[55] D. P. Curran, L. H. Kuo, Tetrahedron Lett. 1995, 36, 6647 6650.
[56] D. P. Curran, L. H. Kuo, J. Org. Chem. 1994, 59, 3259 3261.
[57] T. Schuster, M. Kurtz, M. W. Gˆbel, J. Org. Chem. 2000, 65, 1697
1701.
[75] Y. Kobuke, T. Sugimoto, J. Furukawa, J. Am. Chem. Soc. 1972, 94,
3633 3635.
[76] S. Otto, G. Boccaletti, J. B. F. N. Engberts, J. Am. Chem. Soc. 1998,
120, 4238 4239.
[77] S. Otto, J. B. F. N. Engberts, J. Am. Chem. Soc. 1999, 121, 6798 6806.
[78] Z. Zhu, J. H. Espenson, J. Am. Chem. Soc. 1997, 119, 3507 3512.
[79] T. Loh, J. Pei, M. Lin, Chem. Commun. 1996, 2315 2316.
[80] T. Loh, J. Pei, M. Lin, Chem. Commun. Corrigenda 1997, 505.
[81] E. Keller, B. L. Feringa, Tetrahedron Lett. 1996, 37, 1879 1882.
[82] E. Keller, B. L. Feringa, Synlett 1997, 842 844.
[83] S. Kobayashi, Synlett 1994, 689 701.
[58] A. Wittkopp, P. R. Schreiner, in The Chemistry of Dienes and
Polyenes, Vol. 2 (Ed.: Z. Rappoport), John Wiley & Sons Ltd,
Chichester, 2000, pp. 1029 1088.
[84] S. Kobayashi, Chem. Lett. 1991, 2187 2190.
[85] S. Kobayashi, I. Hachiya, M. Araki, H. Ishitani, Tetrahedron Lett.
1993, 34, 3755 3758.
[59] S. Otto, F. Bertoncin, J. B. F. N. Engberts, J. Am. Chem. Soc. 1996,
118, 7702 7707.
[60] C.-J. Li, Chem. Rev. 1993, 93, 2023 2035.
[61] R. Breslow, U. Maitra, D. Rideout, Tetrahedron Lett. 1983, 24, 1901
1904.
[62] D. C. Rideout, R. Breslow, J. Am. Chem. Soc. 1980, 102, 7816 7817.
[63] W. Blokzijl, J. B. F. N. Engberts, Angew. Chem. 1993, 105, 1610
1648; Angew. Chem. Int. Ed. Engl. 1993, 32, 1545 1579.
[64] J. W. Wijnen, J. B. F. N. Engberts, J. Org. Chem. 1997, 62, 2039 2044.
[65] F. G. Bordwell, D. J. Algrim, J. A. Harrelson, Jr., J. Am. Chem. Soc.
1988, 110, 5903 5904.
[86] S. Kobayashi, Eur. J. Org. Chem. 1999, 15 27.
[87] J. B. F. N. Engberts, Pure Appl. Chem. 1995, 67, 823 828.
[88] S. Otto, W. Blokzijl, J. B. F. N. Engberts, J. Org. Chem. 1994, 59,
5372 5376.
[89] By ™hydrophobicity∫ we mean all interactions not governed by
hydrogen bonding or electrostatics.
[90] R. Breslow, Acc. Chem. Res. 1991, 24, 159 164.
[91] A. Meijer, S. Otto, J. B. F. N. Engberts, J. Org. Chem. 1998, 63, 8989
8994.
[92] D. A. Stauffer, R. E. Barrans, D. A. Dougherty, J. Org. Chem. 1990,
55, 2762 2767.
[66] H.-D. Jakubke, H. Jeschkeit, Concise Encyclopedia Chemistry,
de Gruyter, Berlin, 1994.
[93] Y.-H. Tsao, D. F. Evans, H. Wennerstrˆm, Science 1993, 262, 547
550.
¬
[67] N. C. Hodge, P. Y. S. Lam, C. J. Eyermann, P. K. Jadhav, Y. Ru, C. H.
Fernandez, G. D. De Lucca, C.-H. Chang, R. F. Kaltenbach, E. R.
Holler, F. Woerner, W. F. Daneker, G. Emmett, J. C. Calabrese, P. E.
Aldrich, J. Am. Chem. Soc. 1998, 120, 4570 4581.
[68] R. Huisgen, Pure Appl. Chem. 1980, 52, 2283 2302.
[69] G. Steiner, R. Huisgen, J. Am. Chem. Soc. 1973, 95, 5056 5058.
[70] P. R. Schreiner, A. Wittkopp, 2002, unpublished results.
[71] S. Otto, J. B. F. N. Engberts, J. C. T. Kwak, J. Am. Chem. Soc. 1998,
120, 9517 9525.
[94] K. B. Simonsen, P. Bayon, R. G. Hazell, K. V. Gothelf, K. A.
Jorgensen, J. Am. Chem. Soc. 1999, 121, 3845 3853.
[95] C. S. Wilcox, E. Kim, D. Romano, L. H. Kuo, A. L. Burt, D. P.
Curran, Tetrahedron Lett. 1995, 51, 621 634.
[96] F. G. Bordwell, W.-Z. Liu, J. Am. Chem. Soc. 1996, 118, 8777 8781.
[97] W. G. Hollis, P. L. Smith, D. K. Hood, S. M. Cook, J. Org. Chem.
1994, 59, 3485 3486.
[98] M. Jayamani, N. Pant, S. Ananthan, K. Narayanan, C. N. Pillai,
Tetrahedron 1986, 42, 4325 4332.
[72] K. N. Houk, R. W. Strozier, J. Am. Chem. Soc. 1973, 95, 4094 4096.
[73] J. Sauer, J. Kredel, Tetrahedron Lett. 1966, 7, 731 736.
[74] Y. Kobuke, T. Fueno, J. Furukawa, J. Am. Chem. Soc. 1970, 92, 6548
6553.
[99] K. Shibata, I. Katsuyama, M. Matsui, H. Muramatsu, J. Heterocycl.
Chem. 1991, 28, 161 165.
[100] C. Kloetzel, J. Am. Chem. Soc. 1957, 79, 4226 4227.
Received: July 11, 2002 [F4248]
414
¹ 2003 WILEY-VCH VerlagGmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0414 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2