Self-activated supramolecular reactions: Effects of host-guest recognition on the kinetics of the Diels-Alder reaction of open-chain oligoether quinones with cyclopentadiene

Article abstract of DOI:10.1021/ja021444t

Diels-Alder reactions of acyclic oligoether-substituted quinones 1b, 1c, 2b, and 2c with cyclopentadiene were accelerated by the addition of alkali and alkaline earth metal perchlorates, and scandium trifluoromethane sulfonate (k_c/k_f

Full text of DOI:10.1021/ja021444t

Published on Web 04/19/2003
Self-Activated Supramolecular Reactions: Effects of
Host-Guest Recognition on the Kinetics of the Diels-Alder
Reaction of Open-Chain Oligoether Quinones with
Cyclopentadiene
Akihiko Tsuda,*,† Chikako Fukumoto, and Takumi Oshima*
Contribution from the Department of Materials Chemistry, Graduate School of Engineering,
Osaka UniVersity, Toyonaka, Osaka 560-0043, Japan
Received December 13, 2002; E-mail: tsuda@macro.t.u-tokyo.ac.jp; oshima@ch.wani.osaka-u.ac.jp
Abstract: Diels-Alder reactions of acyclic oligoether-substituted quinones 1b, 1c, 2b, and 2c with
cyclopentadiene were accelerated by the addition of alkali and alkaline earth metal perchlorates, and
scandium trifluoromethane sulfonate (kc/kf ) 1.2-23 for univalent cations, 11-1160 for divalent cations,
and 1700-192 000 for Sc3+, where kc and kf are the rate constants for the metal complexed and
uncomplexed quinones, respectively). The shorter-armed 1a, 2a, and 3, however, exhibited no such
acceleration effects. The rate accelerations can be rationalized by the FMO consequence in which the
bound guest cation withdraws electron density from the quinone dienophile and lowers the LUMO energy
suitable for the orbital interaction with the HOMO of cyclopentadiene. Despite the poor cation selectivity,
these acyclic oligoether quinones showed larger rate accelerations than the relevant quinocrown ethers 4
(kc/kf ) 1.3-3.0 for univalent cations, 5.0-160 for divalent cations, and 100-2020 for Sc3+). The effective
electron withdrawal, which leads to the enhanced rate acceleration, can be caused by the direct interaction
between the metal cation accommodated in the pseudo-cyclic oligoether linkage and the quinone carbonyl
oxygen, as indicated by 1H NMR spectroscopy. In addition, the larger rate enhancement is rather achieved
in the complex with low binding constant K, because the strong encapsulation of metal cation by the
oligoether chain diminishes the crucial interaction to the quinone carbonyl oxygen. As a whole, the smaller
and higher valent cations tend to bring about notable rate acceleration due to the more enhanced ion-
dipole interaction with the quinone carbonyl oxygen. Spectroscopic titration (absorption and 1H NMR) and
kinetic experiments indicated that only the longest di-armed 2c constructs 1:1, and then 1:2, host/guest
complexes with Ca2+, Sr2+, and Ba2+. These 1:2 complexes exhibited the most effective acceleration for
the respective metal cations.
Introduction
supramolecules also enhances their reactivity due to the close
location and independence from external environments (B).3
Catalysis of organic reactions by supramolecules is a current
research interest in supramolecular chemistry.1 These catalytic
reactions can be divided into about three categories as repre-
sented in Scheme 1. The guest-activation by molecular recogni-
tion, in which the electronic properties and environment of the
guest reactant are altered by complexation, brings about
acceleration of the reactions (A).2 The close-packing by ac-
commodation of both reactants into the space or cavity of the
However, little is known about “self-activation” of host
molecules by incorporation of the guest (C).4,5
In organisms, it is well known that metal ion recognition by
proteins plays an important role as a trigger of reactions related
with many biological phenomena.6 In those systems, proteins
(3) (a) Kang, J.; Rebek, J., Jr. Nature 1997, 385, 50. (b) Endo, K.; Koike, T.;
Sawaki, T.; Hayashida, O.; Masuda, H.; Aoyama, Y. J. Am. Chem. Soc.
1997, 119, 4117. (c) Kang, J.; Santamaria, J.; Hilmersson, G.; Rebek, J.,
Jr. J. Am. Chem. Soc. 1998, 120, 3650. (d) Kang, J.; Santamaria, J.;
Hilmersson, G.; Rebek, J., Jr. J. Am. Chem. Soc. 1998, 120, 7389. (e) Clyde-
Watson, A.; Vidal-Ferran, A.; Twyman, L. J.; Walter, C. J.; McCallien, D.
W. J.; Fanni, S.; Bampos, N.; Wilie, R. S.; Sanders, J. K. L. New J. Chem.
1998, 22, 493. (f) Bassani, D. M.; Darcos, V.; Mahony, S.; Desvergne,
J.-P. J. Am. Chem. Soc. 2000, 122, 8795. (g) Yoshizawa, M.; Takeyama,
Y.; Kusukawa, T.; Fujita, M. Angew. Chem., Int. Ed. 2002, 41, 1347.
(4) (a) Inoue, Y.; Ouchi, M.; Hayama, H.; Hakushi, T. Chem. Lett. 1983, 431.
(b) Cacciapaglia, R.; Lucente, S.; Mandolini, L.; Doorn, A. R.; Reinhoudt,
D. N.; Verboom, W. Tetrahedron 1989, 45, 5293. (c) Cacciapaglia, R.;
Doom, A. R.; Mandolina, L.; Reinhoudt, D. N.; Verboon, W. J. Am. Chem.
Soc. 1992, 114, 2611. (d) Itoh, S.; Taniguchi, M.; Takada, N.; Nagatomo,
S.; Kitagawa, T.; Fukuzumi, S. J. Am. Chem. Soc. 2000, 122, 12087.
(5) (a) Tsuda, A.; Oshima, T. New J. Chem. 1998, 1027. (b) Tsuda, A.; Oshima,
T. J. Org. Chem. 2002, 67, 1282.
† Present address: Department of Chemistry and Biotechnology, School
of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo
113-8656, Japan.
(1) Lehn, J-.M. Supramolecular Chemistry; VCH: Weinheim, Germany, 1995.
(2) (a) Hosseini, M. W.; Lehn, J.-M. J. Am. Chem. Soc. 1987, 109, 7047. (b)
Diederich, F.; Lutter, H.-D. J. Am. Chem. Soc. 1989, 111, 8438. (c) Gobbi,
A.; Landini, D.; Maria, A.; Penso, M. J. Chem. Soc., Perkin Trans. 2 1996,
2505. (d) Mattei, P.; Diederich, F. HelV. Chim. Acta 1997, 80, 1555. (e)
Cacciapaglia, R.; Mandolini, L.; Arnecke, R.; Bo¨hmer, V.; Vogt, W. J.
Chem. Soc., Perkin Trans. 2 1998, 419. (f) Brrecia, P.; Cacciapaglia, R.;
Mandolini, L.; Scorsini, C. J. Chem. Soc., Perkin Trans. 2 1998, 1257. (g)
Reetz, M. T.; Waldvogel, S. R. Angew. Chem., Int. Ed. Engl. 1997, 36,
865. (h) Castelli, V. v. A.; Cort, A. D.; Mandolini, L. J. Am. Chem. Soc.
1998, 120, 12688. (i) Merlau, M. L.; Mejia, M. d. P.; Nguyen, S. T.; Hupp,
J. T. Angew. Chem., Int. Ed. 2001, 40, 4239.
(6) Borle, A. B. ReV. Physiol. Biochem. Pharmacol. 1990, 90, 13.
9
10.1021/ja021444t CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 5811-5822
5811

A R T I C L E S
Tsuda et al.
Scheme 1. Schematic Representations of Supramolecular Catalytic Reactionsa
a (A) Guest-activation catalysis, (B) close-packing catalysis, and (C) host-activation catalysis.
Chart 1. Acyclic Oligoether Quinones 1, 2, Reference 3,
Quinocrown Ethers 4, and Metal Cations Used in this Study
Figure 1. Schematic representation of (a) a helical complex and (b)
S-shaped dinuclear complexes.
extent of interaction between the incorporated guest cation and
oxygen atoms attached at the 2,3-positions of quinone governs
the magnitude of rate acceleration. Thus, the complex structure
that can enforce close location between the ring-bound cation
and those oxygens is highly effective in causing large electron
withdrawal from the quinone moiety. Hence, the size-fitted
complexes, which may always keep a close distance between
the cation and the quinone, can bring about larger acceleration
than other size-mismatched complexes. (3) Valency of the
cation: the increase of Lewis acidity with increasing valency
of the cation leads to the significant rate enhancement.
These interesting findings prompted us to extend the cation-
recognized DA reactions to acyclic mono-armed and di-armed
oligoether quinones 1 and 2 (Chart 1), because the oligoether
chain can accommodate the metal cations with characteristic
pseudo-cyclic “helical” conformations, whose structure may lead
to effective separation of the incorporated metal cation from
its counteranions and/or solvent molecules (Figure 1a).8,9 As
another interesting behavior, the long oligoether chains can
construct multinuclear complexes such as 1:2 ligand/metal
complexes with S-shaped manner (Figure 1b).10-12 Their
intrinsic flexibilities have bestowed some promising unpredict-
recognize the appropriate metal cations of peculiar size, valence,
stiffness, and concentration to start certain reactions programmed
for the respective combinations. The metal cations, therefore,
act as a signal material for the information transfer in biological
systems. Hence, construction of artificial self-activation models,
which mimic such biological systems, will become one way to
explore the complicated reaction systems in nature.
We previously reported the effects of host-guest recognition
on the kinetics of the Diels-Alder (DA) reaction of quinone
crown ethers 4 with cyclopentadiene, in which cation binding
of the quinocrown ethers caused large rate accelerations
especially in the size-favorable combinations.5 The principle of
the rate acceleration of DA reactions can be readily rationalized
by assuming that the incorporated guest cation behaves as an
electron-withdrawing group for the quinone dienophile, thus
lowering its LUMO energy against the HOMO of cyclopenta-
diene.7 In that study, we have found that the magnitude of
acceleration is mainly determined by the following three factors.
(1) Complex formation stability (binding constant KQMm+): the
concentration of the complex affects the magnitude of rate
enhancement. (2) Coordination geometry of the complex: the
(8) (a) Vo¨gtle, F.; Weber, E. Angew. Chem., Int. Ed. Engl. 1979, 18, 753. (b)
Saenger, W.; Brand, H. Acta Crystallogr., Sect. B 1979, 35, 838. (c) Weber,
G.; Saenger, W.; Vo¨gtle, F.; Sieger, H. Angew. Chem., Int. Ed. Engl. 1979,
18, 226.
(9) (a) Suzuki, Y.; Morozumi, T.; Nakamura, H.; Shimomura, S.; Bartsh, R.
A. J. Phys. Chem. B 1998, 102, 7910. (b) Morozumi, T.; Anada, T.;
Nakamura, H. J. Phys. Chem. B 2001, 105, 2923.
(10) Iwamoto, R. Bull. Chem. Soc. Jpn. 1973, 46, 1114.
(11) Weber, G.; Saenger, W. Angew. Chem. 1979, 91, 237; Angew. Chem., Int.
Ed. Engl. 1979, 18, 227.
(7) (a) Woodward, R. B.; Hoffmann, R. Angew. Chem. 1969, 81, 797. (b)
Fleming, I. Frontier Orbitals and Organic Chemical Reactions; John
Wiley: London, 1976.
(12) Arunasalam, V.-C.; Baxter, I.; Drake, S. R.; Hursthouse, M. B.; Malik, K.
M. A.; Miller, S. A. S.; Mingos, D. M. P.; Otway, D. J. J. Chem. Soc.,
Dalton Trans. 1997, 1331.
9
5812 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003

Self-Activated Supramolecular Reactions
A R T I C L E S
able structures on the complexation with metal cations. How-
ever, they have displayed weak affinities with guest cations in
comparison with crown ethers.13 The conformational entropy
change for complexation is more favorable for the cyclic ligands,
because their donor oxygens are preorganized to interact with
a guest species, whereas the acyclic ligands must lose significant
freedom to adopt a favorable cyclic conformation for guest
binding. With these backgrounds, the quinones bearing an
oligoether chain were expected to show interesting behavior for
the rate enhancement of the DA reactions by reflecting their
unique cation complexation properties.
Herein, we present a systematic study of the effects of cation
recognition on the kinetics of DA reactions of open-chain
oligoether quinones bearing one oligoether sidearm, 1, and two
oligoether sidearms, 2, with cyclopentadiene in comparison with
the reactions of 2,3-dimethoxy-1,4-benzoquinone, 3, as a
reference compound and quinocrown ethers, 4, reported previ-
ously (Chart 1 and Scheme 2).14 The guest cations used in this
Figure 2. Dependence of the absorption spectrum of 1c (0.2 mM) at 270
nm on the concentration of Ba(ClO4)2 (0-0.6 mM) in CH3CN at 303 K.
The Aobsd values are plotted against the molar equivalent of added salt.
study were univalent Li+, Na+, K+, Rb+, Cs+, divalent Mg2+
Ca2+, Sr2+, Ba2+, and trivalent Sc3+
,
270 nm by addition of Ba(ClO4)2 into the MeCN solution
containing mono-armed 1c (0.2 mM) at 303 K is shown in
Figure 2, where the absorbance of 1c at λmax ) 270 nm increases
with increasing addition of metal salt. The binding constants
.
Scheme 2. Diels-Alder Reaction of the 2c‚Mm+ Complex with
Cyclopentadiene
m+
KQM were determined as the following equation.
KQMm+ ) [QMm+]/[Q][Mm+] )
(Af - Aobsd)/(Aobsd - Ac)[Mm+] (4)
Results and Discussion
where
Complex Formation between Acyclic Oligoether Quinone
and Cation. 1:1 Host-Guest Complexation. The DA reactions
in the presence of a metal ion, in which the acyclic host and
cation are supposed to form a complex with 1:1 stoichiometry,
are to be considered as the following eqs 1-3. In eq 1, the
metal ion is able to form a 1:1 complex with the acyclic
[Mm+] ) [Mm+]t - [Q]t(Af - Aobsd)/(Af - Ac)
(5)
The Ac values can be obtained from the observed absorbance
Aobsd values at the point of the large ratio of [Mm+]t to [Q]t,
where Ac is the absorbance of the complex of acyclic oligoether
quinone with the cation, Af is that of the free one, and [Q]t and
[Mm+]t are the total concentrations of the quinone and the metal
ion, respectively. Under these conditions, [Mm+] saturates
against [Q]; therefore, the Aobsd obtained can be regarded as
oligoether quinone Q, where reaction 1 is at rapid equilibrium
m+
with its binding constant KQM
.
Ac.5b,17 Using this Ac value, we calculated the KQM in eq 4,
m+
and the reasonably reliable values of Ac and KQMm+ can also be
obtained by nonlinear least squares curve-fitting analysis.15 The
obtained 1:1 binding constants for divalent metal cations are
compiled in Table 2.
The [2+4] cycloadducts were formed from the DA reaction
of free quinone (path 2) or a metal-complexed quinone (path
3) with cyclopentadiene, being characterized by the rate
coefficients kf and kc, respectively. The observed rate constant
kobsd is expressed as eq 6.
(14) It is well known that Diels-Alder reactions of quinones with cyclopen-
tadiene give the [4+2] endo adducts.5,28b The reactions of 1c with
cyclopentadiene in the presence or absence of added metal salts, for
example, also afforded endo adducts quantitatively. Adduct of 1c: 1H NMR
(270 MHz, CDCl3) δ ) 1.43 (m, 1H, CP), 1.53 (m, 1H, CP), 3.23 (t, J )
2.7, 2H, CP), 3.38 (s, 3H, Me), 3.53-3.57 (m, 4H, ether), 3.61-3.72 (m,
6H, ether and Q), 3.83-3.86 (m, 2H, ether), 3.94-3.98 (m, 2H, ether),
5.89 (s, 1H, Q), and 6.07 (m, 2H, CP); IR (KBr) 2925, 1694, 1654, 1601,
1458, 1352, 1238, 1107, 867, 728, and 489 cm-1; FAB-MS m/z 335, calcd
for C18H22O6, 334. Anal. Calcd for C18H22O6: C, 64.66; H, 6.63. Found:
C, 64.73; H, 6.81.
m+
The binding constants KQM for complexation of acyclic
oligoether quinones with metal cations were estimated by the
titration experiments of UV-visible spectroscopic change due
to the complexation of the oligoether quinone with the cat-
ion.15,16 The titration profile of a typical absorption change at
(15) (a) Leggett, D. J. Computational Methods for the Determination of
Formation Constants; Plenum: New York, 1985. (b) Connores, K. A.
Binding Constants; Wiley: New York, 1987.
(16) Benesi, H.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71, 2703.
(17) Aoyama, Y.; Asakawa, M.; Matsui, Y.; Ogoshi, H. J. Am. Chem. Soc. 1991,
113, 6233.
(13) (a) Izatt, R. M.; Eatough, D. J.; Christensen, J. J. Struct. Bonding 1973,
16, 161. (b) Frensdroff, H. K. J. Am. Chem. Soc. 1971, 93, 600. (c) Cram,
D. J.; Cram, J. M. Acc. Chem. Res. 1978, 11, 8. (d) Kyba, E. P.; Helgeson,
R. C.; Madan, K.; Gokel, G. W.; Tarnowski, T. L.; Moor, S. S.; Cram, D.
J. J. Am. Chem. Soc. 1977, 99, 2564.
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5813

A R T I C L E S
Tsuda et al.
Table 1. Rate Accelerations of Diels-Alder Reactions of 1, 2, and
3 with Cyclopentadiene in the Presence or Absence of Added
Alkali Metal Perchlorates in Acetonitrile at 30 °Ca
Table 2. Rate Accelerations of Diels-Alder Reactions of 1, 2, 3,
and 4 with Cyclopentadiene in the Presence of Alkaline Earth
Metal Perchlorate in Acetonitrile at 30 °C and Equilibrium and
Rate Constantsa
c
k
obsd/k for each additive
102 k/
f
f
quinone
cationb
kc/M-1 s-1 c
log KQM
kc/k
f
2+
+
quinone
M-1 s-1 b
Li+
Na+
K
Rb+
Cs+
1a
Mg2+
Ca2+
Sr2+
d
d
d
d
d
d
d
d
1a
1b
1c
2a
2b
2c
3
2.02
2.23
2.25
4.74
4.57
4.32
5.33
1.13
1.07
1.20
1.22
1.80
3.68
1.03
1.11
1.18
1.92
1.18
1.39
3.45
1.00
1.09
1.06
1.17
1.14
1.17
1.85
1.04
1.09
1.16
1.16
1.08
1.15
1.74
1.04
1.01
1.20
1.23
1.03
1.05
1.39
1.02
Ba2+
Mg2+
Ca2+
Sr2+
1b
1c
2a
2b
2c
d
d
6.94
1.79
0.84
8.69
7.35
1.83
0.88
d
d
d
d
d
6.85
1.73
0.87
50.2
4.13
15.4
1.70
4.65
0.48
3.51
d
2.45
2.14
2.12
0.79
3.13
3.39
3.70
d
d
d
d
d
3.53
3.03
3.57
2.38
5.09
1.97
4.49
2.01
5.75
1.33
d
311
80.3
37.7
386
327
82.1
39.5
Ba2+
Mg2+
Ca2+
Sr2+
a Kinetic reactions were carried out under pseudo-first-order conditions
by using a large excess of cyclopentadiene (17.0 mM) with respect to
quinones (0.17 mM). The observed second-order rate constants kobsd were
the average of at least two measurements. The error limit of the kobsd is
(2%. b The kf values are taken from the metal-free reaction. c The 24 equiv
excess of alkali metal perchlorate (4.0 mM) with respect to quinones was
used. The counteranion perchlorate was omitted.
Ba2+
Mg2+
Ca2+
Sr2+
Ba2+
Mg2+
Ca2+
Sr2+
kf + kcKQMm+[Mm+
1 + KQMm+[Mm+
]
150
37.9
19.0
1162
95.6
356
39.4
108
kobsd
)
(6)
]
Ba2+
Mg2+
Ca2+
(1:1)
(1:2)
(1:1)
(1:2)
(1:1)
(1:2)
As a similar procedure for the estimation of Ac, the rate constants
kc can also be obtained from the kobsd values at the saturated
concentration of [QMm+], which is confirmed by the kinetic
experiments with varying cation concentration. The produced
[4+2] adducts (A) may have little effect on binding constants
(KQMm+ = KAMm+), because the observed pseudo-first-order rate
constants were not affected during the reactions by their adduct
formation even in the various concentrations of the metal ion
(an excellent correlation coefficient is always attained, r >
0.999).
Sr2+
Ba2+
11.1
81.3
3
Mg2+
Ca2+
Sr2+
d
d
d
d
d
d
Ba2+
Mg2+
Ca2+
Sr2+
4ae
4be
4ce
0.44
0.62
0.51
0.32
d
0.34
0.56
0.47
7.71
0.81
0.25
0.35
2.65
4.11
4.72
4.62
d
5.28
5.33
6.33
2.11
4.93
5.02
5.64
8.4
12.0
9.8
Ba2+
Mg2+
Ca2+
Sr2+
6.2
For most of the combinations of acyclic oligoether quinones
and metal cations, the binding constants KQM and the rate
m+
7.3
12.2
10.2
162
17.0
5.3
constants kc of DA reactions could be calculated by the above
procedures. Therefore, complexation of the acyclic oligoether
quinones with the cations proceeds mainly with 1:1 stoichiom-
etry despite the large flexibility of the open-chain oligoethers.
However, formations of 1:2 host-guest complexes were
confirmed in the largest quinone 2c.
Ba2+
Mg2+
Ca2+
Sr2+
Ba2+
7.3
a Kinetic reactions were carried out under pseudo-first-order conditions
by using 10-100 equiv excess of cyclopentadiene (2.0-17.0 mM) with
respect to quinones (0.17 mM). b Counteranion perchlorate is omitted. c The
kc values are obtained from kobsd at the saturation point of [Q‚M2+]. The
observed second-order rate constants kobsd were the average of at least two
measurements. The error limit of the kobsd is (2%. d No applicable
absorption change and rate acceleration to calculate log KQM2+ and kc were
observed. e These values were used from our previous paper (ref 5b).
1:2 Host-Guest Complexation. Although most of the
combinations of quinones and cations constructed the 1:1
complexes in good accordance with the above equations,
additions of divalent perchlorates such as Ca2+, Sr2+, and Ba2+
into the CH3CN solution of acyclic ligand 2c resulted in a
significant change in the UV-visible spectrum to afford profiles
which were clearly different from 1:1 complexation (Scheme
2). For example, the complexation of 2c with Ca2+ brought
about an absorption increase at around 273 nm, and the plot of
the increase for the concentration of the added metal salt
provided a profile having two saturation points (Figure 3a). Such
a spectroscopic change until the first saturation may be
interpreted as formation of the 1:1 2c‚Ca2+ complex, and the
absorption increase by further addition of Ca2+ salt until the
The much larger K11 value which disregards the 1:2 com-
plexation in eq 7 can be calculated by the procedure employed
for 1:1 complexation as described above. The K12 value in eq
8 can also be determined by using this K11 value as follows.18
The absorbance Aobsd at 273 nm of the complex 2c‚(Ca2+)2 can
be expressed by eq 9, where A0 and A1 are the absorbances of
free 2c and of the 1:1 complex 2c‚Ca2+, respectively, and A2 is
the absorbance of the 1:2 complex 2c‚(Ca2+)2 in the presence
of a large excess of Ca2+. Equation 9 can be rewritten as eq
10. Because the value of the left-hand side in eq 10 is obtained
by using the K11 value, the K12 value can be determined from
the slope of a linear plot between the left-hand side in eq 10
and (A2 - Aobsd)[Ca2+] (Figure 3b). From this procedure, we
second saturation may be attributable to the 1:2 2c‚(Ca2+
complex.
)
2
K11
Q + Mm+ y z QMm+
(7)
(8)
K12
QMm+ + Mm+ y z Q(Mm+
)
2
(18) Fukuzumi, S.; Kondo, Y.; Mochizuki, S.; Tanaka, T. J. Chem. Soc., Perkin
Trans. 2 1989, 1753.
9
5814 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003

Self-Activated Supramolecular Reactions
A R T I C L E S
Figure 3. (a) Dependence of the absorption spectrum of 2c (0.2 mM) at 278 nm on the concentration of Ca(ClO4)2 (0-50 mM) in CH3CN at 303 K. (b)
Plot of ((1 + K[Ca2+])Aobsd - A0)/K[Ca2+] versus (A2 - Aobsd)[Ca2+]. Observed slope ) 85.1, intercept ) 0.60, and r2 ) 0.986.
could obtain the binding constants of log K11 ) 5.09 and log
K12 ) 1.93 for the 1:1 and 1:2 complexes 2c‚Ca2+ and
2c‚(Ca2+)2, respectively. The obtained 1:2 binding constants K11
for the other divalent Sr2+ and Ba2+ are also listed in Table 2.
quinone carbonyl groups actually participates in the cation
binding by judging from the enhanced downfield shifts at the
vinyl protons. The magnitudes of the downfield shifts in the
order of 3-position (∆δ ) 0.17 ppm) > 6-position (0.15) >
5-position (0.14) are in harmony with the efficiency in the
cation-induced electron withdrawal through the π-resonate
carbonyl group (1-position) and the adjacent alkoxy linkage (2-
position).
A0 + K11[Mm+]A1 + K11K12[Mm+]2A2
Aobs
)
(9)
1 + K11[Mm+] + K11K12[Mm+ 2
]
Interestingly, di-armed 2c displayed two-step saturation with
Ba2+, which corresponds to the formations of 1:1 and 1:2
complexes, as is also observed in the UV/vis titration experi-
ments of the absorption spectrum (Supporting Information). At
the first saturation, which was achieved by the addition of ca.
1 equiv of Ba2+, larger downfield shifts were observed for the
two oligoether moieties (∆δ ) 0.17-0.23 ppm) relative to the
quinone protons (0.12 ppm) (Scheme 3). The quinone protons
of 2c‚Ba2+ exhibited a larger downfield shift by 0.04 ppm than
those of 4b‚Ba2+, but a smaller shift by 0.02-0.03 ppm than
those of 1c‚Ba2+. These observations and its large binding
(1 + K11[Mm+])Aobs - A
0 ) K12(A2 - Aobs)[Mm+] + A1
(10)
K11[Mm+
]
With respect to the kobsd under the formation of 1:2 complex
2c‚(M2+)2, we can express them as being similar to eqs 9 and
10, in which the absorbance A is altered to the rate constant k,
because all of the kinetic profiles obtained were fundamentally
in agreement with the respective UV-visible titration profiles.
Plausible Complex Structures Based on 1H NMR Spectral
Analysis. The complexations of cyclic or acyclic oligoethers
with metal cations generally bring about downfield shifts of
2+
constant (log KQM ) 5.75) suggest that 2c preferentially
accommodates Ba2+ in the highly flexible pseudo-cyclic cavity
constructed by the two oligoether chains probably with a little
participation of the carbonyl oxygen to the complexations. At
the second saturation which was attained by the addition of ca.
200 equiv of Ba2+, a larger downfield shift from the 1:1
complexation was noticed for the quinone protons (∆δ ) 0.15
ppm), although small shifts were observed for the oligoether
moiety (-0.05-0.03 ppm). This enhanced downfield shift at
quinone strongly indicates formation of 1:2 complex 2c‚(Ba2+)2,
where two carbonyl oxygens take part in the cation binding
(Scheme 3).20 In fact, this total chemical shift at quinone (∆δ
) 0.27 ppm) is well consistent with the expected value (∆δ )
0.29 ppm) for the bis-lariat complex by summation of the
relevant values of 1c‚Ba2+ (0.15 and 0.14 ppm for 5- and
6-positions, respectively). However, such 1:2 binding behavior
1
the H NMR signals of the host molecules, whose magnitudes
depend mainly on the geometrical location of the cation in the
complex produced.19 As clear examples, we investigated the
complex structures of 1c‚Ba2+ and 2c‚Ba2+ in comparison with
the size-fitted quinocrown ether complex 4b‚Ba2+ as a reference
1
in the H NMR spectroscopy.
The reference 4b‚Ba2+ exhibited the larger downfield shifts
at the crown ether moiety (∆δ ) 0.18-0.27 ppm) than at the
vinyl protons of quinone (0.08 ppm) (Scheme 3). This is because
the incorporated metal cation can directly interact with the crown
ether oxygen atoms, but indirectly exerts the Lewis acidity to
the remote part of the quinone.
In contrast, mono-armed 1c‚Ba2+ showed comparable larger
downfield shifts for both the oligoether ethylene (∆δ ) 0.16-
0.21 ppm) and the quinone vinyl protons (0.14-0.17 ppm).
These NMR chemical shifts indicate formation of the lariat-
shaped complex as shown in Scheme 3, in which one of the
(20) (a) Echegoyen, L.; Kaifer, A.; Durst, H.; Schultz, R. A.; Dishong, D. M.;
Goli, D. M.; Gokel, G. W. J. Am. Chem. Soc. 1984, 106, 5100. (b) Schultz,
R. A.; White, B. D.; Dishong, D. M.; Arnold, D. K.; Gokel, G. W. J. Am.
Chem. Soc. 1985, 107, 6659. (c) Miller, S. R.; Gustowski, D. A.; Gokel,
G. W.; Echegoyen, L.; Kaifer, A. E. Anal. Chem. 1988, 60, 2021.
(19) Ikeda, A.; Shinkai, S. Chem ReV. 1997, 97, 1713.
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5815

A R T I C L E S
Tsuda et al.
Scheme 3. Observed Shifts (ppm) of 1H NMR (500 MHz) Signals of 1c, 2c, and 4b by Addition of Ba(ClO4)2 in CDCl3 at 25 °C and Their
Possible Complexation Structures Suggested from these Shifts
was extremely reduced in 2b (Supporting Information), in which
ation ability. The experiments were carried out in MeCN
solution containing an acyclic oligoether quinone (0.17 mM)
with a large excess of cyclopentadiene (17 mM) with or without
added 24 equiv excess of alkali metal perchlorate (4 mM) at
303 K.
its respective shorter sidearms cannot strongly bind Ba2+ as
expected from mono-armed 1b‚Ba2+
.
In the following sections, we will discuss the effects of cation
binding on DA reactions of acyclic oligoether quinones with
cyclopentadiene on the basis of complexation features described
above.
In the absence of metal perchlorate, DA reactions of mono-
armed quinones 1a-c, di-armed 2a-c, and reference 3 with
Measurements of Reaction Rates. The kinetic experiments
were performed in an acetonitrile solution containing a quinone
(0.17 mM) and a large excess of cyclopentadiene (2.0-16.0
mM) with or without addition of alkali, alkaline earth metal
perchlorates, or scandium triflate at 303 K. The rates of reactions
were determined by monitoring the disappearance of absor-
bances due to quinones 1a-c at λmax ) 360 nm, and 2a-c and
3 at λmax ) 391-399 nm. The rates obeyed pseudo-first-order
kinetics up to at least two half-lives, and the second-order rate
constants were obtained by dividing the observed first-order rate
constants by the corrected concentration of cyclopentadiene on
the consumption of a half-amount of quinone. In the absence
of metal salts, the second-order rate constants varied in the range
of 2.0-2.3 × 10-2 M-1 s-1 for 1a-c, and 4.3-5.3 × 10-2
M-1 s-1 for 2a-c and 3. The addition of metal salts brought
about more or less the rate enhancement depending on the
combination of quinones and metal cations. The details will be
described for the univalent, divalent, and trivalent metal
complexations, respectively.
cyclopentadiene gave the rate constants kf ) 2.0-2.2 × 10-2
,
4.3-4.7 × 10-2, and 5.3 × 10-2 M-1 s-1, respectively. The
lower reactivity of 1a-c, about one-half that of 2a-c and 3, is
probably due to the reduced inductive effects by the mono-
armed oxygen atom, which would result in the poor drop of
the quinone LUMO energy. In the presence of univalent metal
salts, the quinones 1c, 2b, and 2c bearing relatively longer
oligoether chains undergo more or less the selective rate
enhancement of [4+2] cycloaddition reflecting their character-
istic complexations (Table 1). The overall kinetic features
associated with the cation binding can be more explicitly
visualized in the plots of ln(kobsd/kf) versus metal ion radius
(Figure 4). The most selective acceleration for each quinone
was attained for the complexes 1c‚Na+ (kobsd/kf ) 1.9), 2b‚Li+
(1.8), and 2c‚Li+ (3.7). Contrary to these long-armed quinones,
short-armed quinones 1a, 1b, 2a, and reference 3 showed no
appreciable rate dependency on the addition of metal salts
because of their weak cation affinities. It was also noted that
the conspicuous selective accelerations, which were predicted
from the size-fit concept on the previous studies of crown ethers,
were not observed for both families of quinones 1 and 2.5,21
Larger acceleration seems to be achieved in the combinations
of larger oligoether quinones and smaller cations.
Univalent Cation Recognition. Initially, we performed a
systematic kinetic experiment for DA reactions accelerated by
the univalent cation recognition with the fixed concentration
of host quinone and guest metal ion because of poor complex-
9
5816 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003

Self-Activated Supramolecular Reactions
A R T I C L E S
Figure 5. Dependence of kobsd on the concentration of Ca(ClO4)2 (0-45
mM) for the Diels-Alder reaction of 2c (0.17 mM) with cyclopentadiene
(2.0 mM) in CH3CN at 303 K.
Figure 4. Plot of ln(kobsd/kf) versus the alkali metal ion radius (Å), where
kobsd and kf represent the rate constants for the reactions in the presence
and absence of added alkali metal perchlorates (4.0 mM), respectively. Metal
Divalent Cation Recognition. Effects of Binding Constants,
Geometry, and Stoichiometry. The complexation of quinones
1b, 1c, 2b, and 2c with alkaline earth metal ions brought about
much larger rate accelerations (kc/kf ) 11-1162) as compared
with the univalent alkali metal complexes or size-fitted
quinocrown ether complexes such as 4a‚Ca2+ (kc/kf ) 12.0) and
4b‚Ba2+ (10.2) (Table 2).5 However, the short-armed quinones
1a and 2a as well as reference 3 showed no applicable
acceleration due to their weak cation affinities (kobs/kf ) 1.0-
1.4 even in the presence of 10 equiv of metal cations). The
ion radii (Å): Li+ ) 0.60, Na+ ) 0.95, K+ ) 1.33, Rb+ ) 1.48, Cs+
)
1.69.
For 2c‚Li+ and 2c‚Na+ complexes, which exhibited larger
obsd/kf, the binding constants with 1:1 stoichiometry and the
k
rate constants kc were determined from the titration profiles of
the absorption spectra and the concentration-dependent kinetic
profiles of DA reactions, respectively. The 2c‚Li+ complex has
+
a small binding constant of log K2c‚Li ) 1.28 and a larger rate
2+
constant of kc ) 1.03 M-1 s-1 (kc/kf ) 23.4) as compared with
the size-fitted quinocrown complex 4b‚K+ (kc ) 0.17 M-1 s-1
(kc/kf ) 3.7)).5 The pronounced rate enhancement in 2c‚Li+ may
be attributed to the cooperative participation of the quinone
carbonyl oxygen to the metal complexation, by which the bound
cation may exert effective electron-withdrawing ability on the
conjugated CdC bond of quinone (vide infra).22 Similar curve
binding constants KQM were determined by titration experi-
ments of the absorption spectra, and the saturated rate constants
kc of the DA reactions of the complexes with cyclopentadiene
were also estimated by the saturation point on kinetic profiles.
With respect to the complexation stoichiometry, these oligoether
quinones 1b, 1c, and 2b interacted with divalent cations to form
the 1:1 noncovalent complexes, whereas 2c formed the 1:1 and
then 1:2 complexes with Ca2+, Sr2+, Ba2+ as indicated by the
1H NMR and UV/vis spectroscopy. Such 1:2 complexations
were also observed in the kinetic profiles of 2c on varying the
concentration of divalent cations, as represented for Ca2+ (Figure
5). The rate constants for the 1:1 and 1:2 complexes (kc(1:1) and
fitting provided log K2c‚Na ) 1.97 and kc ) 0.28 M-1 s-1 (kc/
+
kf ) 6.39) for the 2c‚Na+ complex. The more favorable
acceleration for 2c‚Li+ than 2c‚Na+ may be explained by the
more effective electron withdrawal by the smallest Li+ cation.
Thus, the smaller kc but larger K2c‚Na of 2c‚Na+ as compared
+
with that of the 2c‚Li+ complex implies that the larger Na+
cation is more favorably accommodated in the pseudo-cyclic
cavity of 1c than Li+, but exerts reduced electron withdrawal
on account of the weaker Lewis acidity.
k
c(1:2)) can be determined from the respective saturation points
in the plots of kobsd versus metal salt concentrations (Table 2).
log KQM2+ versus Metal Ion Radius. A plot of complexation
stabilities (log KQM2+) against the metal ion radius (Å) revealed
some tendencies in the rate enhancements (Figure 6). In all of
These observations indicate that the suitable pseudo-cyclic
complexation of a smaller cation with the aid of a quinone
carbonyl interaction would result in the more enhanced rate
acceleration as deduced from Figure 4. In the next section, the
detailed relationships between rate acceleration and cation
recognition will be described in view of the kc, KQMm+, metal
cation size, and chain length.
2+
the divalent metal complexes, the largest KQM with 1:1
bindings was achieved by 2c. The increased binding constants
from 1b to 1c and from 2b to 2c are also observed for all of
the metal complexes. The more the host moiety enlarges with
increasing oxyethylene units, the more the cation capturing
ability of the acyclic host moiety becomes advantageous.
However, despite the small number of donor oxygens, the mono-
armed 1c, which may construct lariat-shaped complexes by using
five donor oxygens, displays a similar binding profile to di-
armed 2b having six effective oxygens and further can form a
complex with smaller Mg2+. In addition, although 1b and 2a
may have the same number of effective oxygens for complex-
(21) (a) Izatt, R. M.; Bradshaw, J. S.; Nielson, S. A.; Lamb, J. D.; Christensen,
J. J. Chem. ReV. 1985, 85, 271. (b) Bajaj, A. V.; Poonia, N. S. Coord.
Chem. ReV. 1988, 87, 55. (c) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.;
Bruenig, R. L. Chem. ReV. 1991, 91, 1721.
(22) The carbonyl oxygens of quinones have a large affinitiy for metal cations.
Their interactions, where metal cations act as Lewis acid, have been applied
to the catalysis of various organic reactions. (a) Fukuzumi, S.; Okamoto,
T. J. Am. Chem. Soc. 1993, 115, 11600. (b) Vaugeois, J.; Wuest, J. D. J.
Am. Chem. Soc. 1998, 120, 13016.
ations, only mono-armed 1b can form complexes with Ca2+
,
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5817

A R T I C L E S
Tsuda et al.
2+
Figure 6. Plot of log KQM versus ion radius (Å) for the divalent metal
complexes of 1b, 1c, 2b, and 2c (including 1:1 and 1:2). Metal ion radii
(Å): Mg2+ ) 0.65, Ca2+ ) 0.99, Sr2+ ) 1.13, Ba2+ ) 1.35.
Figure 7. Plot of ln(kc/kf) versus ion radius (Å) for the divalent metal
complexes of 1b, 1c, 2b, and 2c (including 1:1 and 1:2).
Sr2+, and Ba2+ with small KQM2+. These observations indicate
that the Lewis basicity of quinone carbonyl oxygen is larger
than that of oligoether oxygens for the complexations. Such a
strong basicity of carbonyl oxygens may also allow for the
formation of 1:2 complexes in 2c.
quinocrown ethers reported previously, the strong accommoda-
tion of the guest cation by the crown ether ring is ineffective
for the rate enhancement, although selective acceleration was
achieved as a result of their selective cation binding (Table 2).5
It should be noted that long-armed 1c and 2c only can bind
Mg2+, bringing about larger rate accelerations. The charge
condensed Mg2+ cation is a potential strong Lewis acid, and
thus the 1c‚Mg2+ complex that is stabilized by the contribution
of quinone carbonyl oxygen, as described above, can produce
larger rate acceleration. In all of the divalent complexes, 2c‚
Mg2+ that exhibits only 1:1 stoichiometry displays the largest
acceleration. The additional sidearm of 2c may be able to assist
more appropriate interactions between Mg2+ and carbonyl
oxygen as compared with 1c, resulting in the largest kc. This
long-arm effect was also observed in the quinocrown 4c‚Mg2+
complex, which displayed the largest acceleration in divalent
In these acyclic oligoether quinones, the cation selectivity is
quite low as compared with that of quinocrown ethers; that is,
the affinity between host and guest is scarcely affected by
variation of the cation size. These acyclic oligoether quinones
having large flexibilities can meet the structural demand from
guest cations for the complexations; in contrast, the quinocrowns
4 cannot accept it due to the restricted flexibility of the cyclic
structures.
ln(kc/kf ) versus Metal Ion Radius. In the plots of ln(kc/kf)
versus metal ion radius in Figure 7, it is found that 1:2 2c‚
(M2+)2 complexes for Ca2+, Sr2+, or Ba2+ ions are the most
effective for the acceleration of DA reactions (Table 2). On the
contrary, acceleration of the corresponding 1:1 2c‚M2+ com-
plexes is less effective for these cations. In comparison of the
rate enhancements of 1:1 complexes, mono-sidearms 1b‚M2+
and 1c‚M2+ exhibit the largest acceleration for Ca2+, Sr2+, and
Ba2+. 1b and 1c have different affinities for these cations, but
provide quite similar kinetic profiles, in which their accelerations
quinocrown complexes 4‚M2+
.
As another observation, the rate accelerations by complex-
ations of acyclic oligoether quinones with divalent cations show
a global tendency in which the smaller cation seems to give a
larger acceleration for all of the oligoether quinones (Figure
7), similar to the case of univalent cations (Figure 4). In all
cases, including 1:2 2c‚(M2+)2 complexes, this tendency can
be observed. This trend may be ascribed to the different Lewis
acidity of the cations.23 There are many reports for the studies
of electrochemical behavior of functionalized quinones bearing
host moieties such as crown and podand in the presence of metal
cations.24 Their complexations were accelerated by the ionic
interaction of reduced quinone with metal cations upon elec-
trochemical reduction. In comparisons of the complex stabilities
of the neutral quinonoid hosts and their reduced forms with
alkali metal cations, the smaller metal cation provided larger
complexation stability in all cases. This observed trend in the
electrochemical behavior of functionalized quinones is attribut-
able to the different charge densities of metal cations and is
are also larger than those of di-sidearms 2b‚M2+ and 2c‚M2+
,
and quinocrown 4‚M2+ complexes. There are probably little
differences among the basic complexation structures of 1b and
1c with Ca2+, Sr2+, and Ba2+; thus, they may adopt the lariat-
shaped complexations as shown in Scheme 3. The larger
contribution of quinone carbonyl oxygen to the complexation
may lead to more effective electron withdrawing from the
quinone moiety, resulting in larger accelerations. The smaller
kc of 2b and 2c as compared with that of 1b and 1c, and further
the more declined kc of 2c than that of 2b, may result from
interference of the interaction between quinone carbonyl oxygen
and chain-bound cation due to the effective encapsulation of
the cation by the two oligoether sidearms. As shown in the
(23) Fukuzumi, S.; Okubo, K. Chem.-Eur. J. 2000, 6, 4532.
9
5818 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003

Self-Activated Supramolecular Reactions
A R T I C L E S
Table 3. Rate Accelerations of Diels-Alder Reactions of 1, 2, 3,
and 4 with Cyclopentadiene in the Presence of Sc(OTf)3 in
Acetonitrile at 30 °C and Equilibrium and Rate Constantsa
kc/M-1 s-1 b
log KQM
kc/k
f
2+ c
quinone
1a
1b
1c
2a
2b
2c
3
d
d
677
4090
d
38.2
8280
(1.46)e
96.2
46.5
4.92
3.2
5.0
14 800
182 000
d
1710
192 000
(27.3)e
2021
977
4.8
4.9
4af
4bf
4cf
2.4
2.7
4.8
103
a Kinetic reactions were carried out under pseudo-first-order conditions
by using 10-100 equiv excess of cyclopentadiene (0.5-5.0 mM) with
respect to quinones (0.05 mM). b The kc values are obtained from kobsd at
the saturation point of [Q‚Sc3+]. The observed second-order rate constants
k
obsd were the average of at least two measurements. The error limit of the
k
obsd is (2%. c Approximate binding constants were calculated from kinetic
profiles. d No applicable rate acceleration to calculate kc was observed.e The
value in parentheses is kobsd for the cycloaddition of 3 (0.17 mM) in the
presence of 24 equiv excess of Sc(OTf)3 (4.0 mM). f These values were
used in our previous paper (ref 5b).
2+
Figure 8. Plot of ln(kc/kf) versus log KQM for the divalent metal
complexes of 1b, 1c, 2b, 2c (including 1:1 and 1:2), and 4a-d.
entirely similar to our study on rate accelerations. Hence, the
combination of a flexible oligoether quinone, which can assist
the metal-carbonyl interaction, and a smaller metal cation with
larger electron density, brings about effective electron with-
drawal associated with the larger acceleration.
Trivalent Sc3+ Recognition. Abnormal Accelerations by
Sc3+ Recognition. Recently, rare-earth metal triflates are well
known as effective catalysts for a variety of organic reactions,
where these metal salts mainly act as a strong Lewis acid.25 In
particular, among them, Sc3+ has displayed excellent catalytic
activity due to its smaller ion size (ion radius ) 0.81 Å) and
hard metal character. For example, oxidative ring-closed reac-
tions of porphyrin oligomers with 2,3-dichloro-5,6-dicyano-p-
benzoquinone (DDQ) proceeded only in the presence of
Sc(OTf)3, where Sc3+ interacts with the quinone carbonyl
oxygens to stabilize the intermediate quinone anion radical.26
The DA reactions of acyclic oligoether quinones 1b, 1c, 2b,
and 2c with cyclopentadiene were dramatically accelerated by
addition of trivalent Sc3+ (kc/kf ) 1710-192 000), although the
kc values for the short-armed 1a, 2a, and 3 could not be
calculated because of the poor host-guest interaction (Table
3). The highest kc/kf value of 192 000 for 2c corresponds to the
ln(kc/kf) versus log KQM2+. The overall kinetic features of
accelerations of DA reactions by divalent cation recognitions
can be more clearly understood in Figure 8, where ln(kc/kf)
values for all of the complexes of 1, 2, and 4 are plotted against
log KQM2+. This plot clarifies a global trend of contribution of
the complex stability to the rate acceleration. As a characteristic
observation, it can be found that the smaller binding constant
of host-guest complexation that can increase interaction
between carbonyl oxygens and the incorporated cation is
necessary to cause larger rate enhancement. Therefore, the
acyclic oligoether quinones, which have larger flexibility as well
as smaller affinities for metal cations, are more advantageous
for rate accelerations than are quinocrown ethers 4. The
restricted cyclic chain in quinocrowns 4 is favorable for selective
cation binding, which is associated with the selective rate
acceleration, but is unfavorable to cause dramatic acceleration.
As an exception, 4c‚Mg2+ bearing larger flexibility with a
smaller binding constant displays the anomalous larger accelera-
tion like oligoether quinones. By following this consideration,
the largest acceleration is expected to be achieved in shorter-
armed 1a, 2a, and reference 3; however, no acceleration was
observed due to their extremely poor binding affinity for all of
the cations. This global tendency for larger acceleration from
Figure 8 indicates the importance of complex structures and
their stabilities for the rate accelerations.
enormous reduction of the activation energy by 30.7 kJ mol-1
,
which amounts to 60% of the uncatalyzed reaction (∆Gq ) 51.6
kJ mol-1). This Sc3+ catalyzed acceleration, which is much
larger than that for quinocrown ethers 4 (kc/kf ) 103-2020),
strongly supporting the participation of quinone carbonyl oxygen
in the lariat complexation. Unfortunately, their reliable binding
constants could not be determined by the titration experiments
in the absorption spectra because of the gradual degradation of
host quinones by added Sc(OTf)3. Hence, approximate binding
constants were calculated from their concentration-dependent
kinetic profiles. The selected profiles of 1c and 2c are shown
in Figure 9, where kobsd/kf values are almost superimposed with
the saturation at approximately 3 equiv of added Sc(OTf)3. In
contrast to the case of divalent cations, there is no indication
of the formation of 1:2 complex as indicated by the plateau
(24) (a) Sugihara, K.; Kamiya, H.; Yamaguchi, M.; Kaneda, T.; Misumi, S.
Tetrahedron Lett. 1981, 22, 1619. (b) Bock, H.; Hierholzer, B.; Vo¨gtle,
F.; Hollmann, G. Angew. Chem., Int. Ed. Engl. 1984, 23, 57. (c) Wolf, R.
E.; Cooper. S. R. J. Am. Chem. Soc. 1984, 106, 4646. (d) Maruyama, K.;
Sohmiya, H.; Tsukube, H. Tetrahedron Lett. 1985, 26, 3583. (e) Gustowski,
D. A.; Degado, M.; Gatto, V. J.; Echegoyen, L.; Gokel, G. W. J. Am. Chem.
Soc. 1986, 108, 7553. (f) Maruyama, K.; Sohmiya, H.; Tsukube, H. J. Chem.
Soc., Perkin Trans. 1 1986, 2069. (g) Deborah, M. D.; Gustowski, D. A.;
Yoo, H. K.; Gatto, V. J.; Gokel, G. W.; Echegoyen, L. J. Am. Chem. Soc.
1988, 110, 119. (h) Togo, H.; Hashimoto, K.; Morihashi, K.; Kikuchi, O.
Bull. Chem. Soc. Jpn. 1988, 61, 3026. (i) Echegoyen, L. E.; Yoo, H. K.;
Gotto, V. J.; Gokel, G. W.; Echegoyen, L. J. Am. Chem. Soc. 1989, 111,
2440.
values even on further addition of Sc3+
.
Absence of 1:2 complex for 2c can be explained by
considering that the 1:1 complexation substantially reduces the
electron density of opposite quinone carbonyl oxygen, and a
possible 1:2 complex, even if formed, must endure an enormous
(25) Kobayashi, S.; Sugiura, M.; Kitagawa, H.; Lam, W. W.-L. Chem ReV. 2002,
102, 2227.
(26) Tsuda, A.; Osuka, A. Science 2001, 293, 79.
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5819

A R T I C L E S
Tsuda et al.
Ca2+, Sr2+, and Ba2+. Application of these 1:2 complexes to
the DA reactions achieved the largest accelerations for all of
the metal cations. As another tendency, the complexations of
acyclic oligoether quinones with a smaller cation brought about
larger rate enhancement, being ascribed to the different charge
densities of the metal cations. In comparison of the uni-, di-,
and trivalent metal complexes, the valence of the cations is found
to be the most effective factor to cause larger rate enhancement.
The self-activated supramolecules are expected to have potential
utility for catalytic reactions, regio-selective reactions, asym-
metric reactions, photonic reactions, and biochemical reactions.
These applied studies will be our next fascinating project.
Experimental Section
General. All reagents and solvents were of the commercial reagent
grade and were used without further purification except where noted.
Dry CH2Cl2 and CHCl3 were obtained by heating under reflux and
distillation over CaH2. Solvents used for spectroscopic measurements
were all spectroscopic grade. Preparative separations were performed
by silica gel gravity-flow column chromatography (Wakogel C-400).
Melting points were measured with a Yanagimoto melting-point
apparatus. 1H NMR spectra were recorded on a JEOL EX-270
spectrometer and Alpha-500 spectrometer in CDCl3, and chemical shifts
were represented as δ values in ppm relative to the internal standard
of trimethylsilane (TMS). IR spectra were recorded on a JASCO FT/
IR-410 spectrometer. FAB mass spectra were recorded on a JEOL HX-
110 spectrometer with the positive-FAB ionization method (accelerating
voltage, 10 kV; primary ion sources, Xe) and 3-nitrobenzyl alcohol
matrix. Elemental analyses were performed on a Yanaco CHN corder
MT-5. UV-visible spectra were recorded on a JASCO V-550
spectrometer equipped with a PELTIER EHC-447 thermostated cell-
holder.
Materials. All of the metal perchlorates and triflate were of
commercial origin and were used without further purification. Spectrally
pure acetonitrile (Nakarai Chemical Co., Ltd.) was used in sample
solutions. The compounds 1a-c and 2b,c were synthesized by
modification of the literature methods.5,27,28 Cyclopentadiene was
purified by distillation just before use. Standard solutions (0.05-1 M)
of cyclopentadiene, which was preserved in a freezer, quinones, and
metal salts were prepared and subsequently used for kinetic experiments.
Kinetic Measurements. The general kinetic experiments of DA
reactions were performed under pseudo-first-order conditions in ac-
etonitrile solution containing a large excess of cyclopentadiene with
respect to quinone at 30 °C as described elsewhere.5b
2-(2′-Methoxyethoxy)-quinone (1a). Sodium hydroxide (2.4 g, 60
mmol) was added to a stirred solution of 2-benzyloxy-phenol 5 (3.0 g,
15 mmol) and monoethyleneglycol monomethyl ether tosylate 6a (3.5
g, 15 mmol) in dry dioxane (150 mL), and then the solution was heated
at 80 °C for 24 h. The precipitate was filtered off, and the filtrate was
evaporated under the reduced pressure to give a viscous residue. The
residue was extracted with benzene to give 2-(2′-methoxyethoxy)-1-
benzyloxybenzene, 7a (3.4 g, 88%). Subsequently, a mixture of 7a and
Pd carbon (1 g) in dioxane (10 mL) was stirred under an atmosphere
of hydrogen for 3 days. The catalyst was filtered off, and the solvent
was evaporated under the reduced pressure to give 2-(2′-methoxy-
ethoxy)phenol, 8a (2.0 g, 91%), as an oil. Fremy’s salt (3.8 g, 14 mmol)
was added to a stirred suspension of 8a (0.55 g, 3 mmol) in 5% aqueous
sodium acetate (280 mL) containing a small amount of benzene (30
mL). After the mixture had been stirred at room temperature for 30
min, it was extracted with benzene (3 × 70 mL). The combined organic
layers were dried (MgSO4) and concentrated under reduced pressure.
Figure 9. Dependence of kobsd on the concentration of Sc(OTf)3 for the
Diels-Alder reaction of 1c and 2c (0.05 mM) with cyclopentadiene (0.5
mM) in MeCN at 303 K. The kobsd values are plotted against the molar
equivalent of added salts.
electrostatic repulsion between the small charge-condensed Sc3+
ions. These considerations may be also applied to the lack of
the 1:2 complexation between 2c and the smallest divalent Mg2+
(0.65 Å).
The observed differences in acceleration in Sc3+ complexes,
therefore, can be explained on the basis of the interaction
between Sc3+ and quinone carbonyl oxygen. The extremely
larger rate enhancement (kc/kf ) 182 000) with larger KQ‚Sc3+
in 1c‚Sc3+ as compared with that in 1b‚Sc3+ (kc/kf ) 14 800)
may be ascribed to the fact that the elongation of one
oxyethylene unit in 1c provides increasing affinity and favorable
configurations for the interaction of Sc3+ and the carbonyl
oxygen in the lariat complexation. The dramatic rate enhance-
ments of 2c‚Sc3+ (kc/kf ) 192 000), thus, can be taken as proving
the direct binding of Sc3+ on the quinone carbonyl oxygen atom
with the aid of two oligoether chains. The drastic decrease of
acceleration in shorter two-armed 2b, which displays smaller
1710-fold acceleration with relatively large KQ‚Sc3+, may result
from the difficult Sc3+-carbonyl interaction in 2b‚Sc3+ due to
the strong encapsulation of Sc3+ into the pseudo-cyclic cavity
constituted of two oligoether sidearms.
Conclusions
The results presented here for the rate acceleration of the
Diels-Alder reaction of metal-complexed oligoether quinones
with cyclopentadiene have demonstrated a development of self-
activated supramolecules. The noncovalent complexes of oli-
goether quinones 1 and 2 with metal cations exhibited larger
reaction rate accelerations with poor selectivity as compared
with those of quinocrown ethers 4. The large electron withdrawal
from the quinone moiety, which is directly associated with the
larger rate accelerations, is brought about by the interaction
between the quinone carbonyl oxygen and the chain-bound
cation. Because strong encapsulation of the metal cation by the
host oligoether moiety prevents the interaction of the cation and
the quinone carbonyl groups, the larger rate enhancement is
achieved in the complexes with weak host-guest interaction.
The two-armed 2c, which has flexible long chains, allowed for
the construction of 1:1 and subsequent 1:2 supramolecules with
(27) Tsuda, A.; Moriwaki, H.; Oshima, T. J. Chem. Soc., Perkin Trans. 2 1999,
1235.
(28) (a) Dietl, F.; Gierer, G.; Merz, A. Synthesis 1985, 626. (b) Hayakawa, K.;
Kido, K.; Kanematu, K. J. Chem. Soc., Perkin Trans. 1 1988, 511.
9
5820 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003

Self-Activated Supramolecular Reactions
A R T I C L E S
The residue was chromatographed on a silica gel with benzene-
chloroform (3:1), and recrystallized from chloroform-hexane to give
1a as yellow needles (303 mg, 55%). 1a: 1H NMR (270 MHz, CDCl3)
δ 3.43 (s, 3H, Me), 3.77-3.81 (m, 2H, oxymethylene), 4.05-4.09 (m,
2H, oxymethylene), 5.96 (s, 1H, quinone), 6.69 (s, 1H, quinone), and
6.70 (s, 1H, quinone); IR (KBr) 2942, 2903, 1693, 1658, 1618, 1591,
2,3-Bis(2′-methoxyethoxy)-quinone (2a). Sodium hydroxide (8.0
g, 200 mmol) was added to a stirred solution of 3-benzyloxypyrocat-
echol 9 (5.0 g, 23 mmol) and monoethyleneglycol monomethyl ether
tosylate 6a (10.6 g, 46 mmol) in dry dioxane (200 mL), and then the
solution was heated at 80 °C for 24 h. The precipitate was filtered off,
and the filtrate was evaporated under the reduced pressure to give a
viscous residue. The residue was extracted with benzene to give 2,3-
bis(2′-methoxyethoxy)-1-benzyloxybenzene, 10a (7.0 g, 91%). Subse-
quently, a mixture of 10a and Pd carbon (1 g) in dioxane (10 mL) was
stirred under an atmosphere of hydrogen for 3 days. The catalyst was
filtered off, and the solvent was evaporated under the reduced pressure
to give 2,3-bis(2′-methoxyethoxy)phenol 11a (4.8 g, 95%) as an oil.
Fremy’s salt (3.8 g, 14 mmol) was added to a stirred suspension of
11a (0.77 g, 3 mmol) in 5% aqueous sodium acetate (280 mL)
containing a small amount of benzene (30 mL). After the mixture had
been stirred at room temperature for 30 min, it was extracted with
benzene (3 × 70 mL). The combined organic layers were dried (MgSO4)
and concentrated under reduced pressure. The residue was chromato-
graphed on a silica gel with chloroform-THF (49:1). After removal
of solvent, 2a was obtained as a red oil (0.25 g, 32%). 2a: 1H NMR
(270 MHz, CDCl3) δ 3.39 (s, 6H, Me), 3.66-3.69 (m, 4H, oxymeth-
ylene), 4.39-4.43 (m, 4H, oxymethylene), and 6.60 (s, 2H, quinone);
IR (KBr) 2928, 1657, 1590, 1127, 1075, and 841 cm-1; FAB-MS m/z
257 (M + 1), calcd for C12H16O6, 256. Anal. Calcd for C12H16O6: C,
57.34; H, 6.29. Found: C, 57.23; H, 6.12. UV/vis (CHCl3): λmax(ꢀ) )
271 (2790) and 399 (1030) nm.
1478, 1460, 1350, 1315, 1238, 1198, 1128, 1103, 1027, and 876 cm-1
;
FAB-MS m/z 183 (M + 1), calcd for C9H10O4, 182. Anal. Calcd for
C9H10O4: C, 59.55; H, 5.41. Found: C, 59.34; H, 5.53. mp 65-66
°C; λmax(ꢀ) ) 358 (1910) nm.
2-(2′-(2′′-Methoxyethoxy)ethoxy)-quinone (1b). Sodium hydroxide
(2.4 g, 60 mmol) was added to a stirred solution of 2-benzyloxyphenol
5 (3.0 g, 15 mmol) and diethyleneglycol monomethyl ether tosylate
6b (4.1 g, 15 mmol) in dry dioxane (150 mL), and then the solution
was heated at 80 °C for 24 h. The precipitate was filtered off, and the
filtrate was evaporated under the reduced pressure to give a viscous
residue. The residue was extracted with benzene to give 2-(2′-(2′′-
methoxyethoxy)ethoxy)-1-benzyloxybenzene, 7b (4.4 g, 92%). Sub-
sequently, a mixture of 7b and Pd carbon (1 g) in dioxane (10 mL)
was stirred under an atmosphere of hydrogen for 3 days. The catalyst
was filtered off, and the solvent was evaporated under the reduced
pressure to give 2-(2′-(2′′-methoxyethoxy)ethoxy)phenol, 8b (2.69 g,
88%), as an oil. Fremy’s salt (3.8 g, 14 mmol) was added to a stirred
suspension of 8b (0.68 g, 3 mmol) in 5% aqueous sodium acetate (280
mL) containing a small amount of benzene (30 mL). After the mixture
had been stirred at room temperature for 30 min, it was extracted with
benzene (3 × 70 mL). The combined organic layers were dried (MgSO4)
and concentrated under reduced pressure. The residue was chromato-
graphed on a silica gel with benzene-chloroform (1:1). After removal
of solvent, 1b was obtained as a yellow oil (0.39 g, 59%). 1b: 1H
NMR (270 MHz, CDCl3) δ 3.38 (s, 3H, Me), 3.54-3.57 (m, 2H,
oxymethylene), 3.69-3.73 (m, 2H, oxymethylene), 3.89-3.92 (m, 2H,
oxymethylene), 4.08-4.11 (m, 2H, oxymethylene), 5.96 (s, 1H,
quinone), 6.69 (s, 1H, quinone), and 6.70 (s, 1H, quinone); IR (KBr)
2883, 1684, 1654, 1591, 1232, 1100, and 871 cm-1; FAB-MS m/z 227
(M + 1), calcd for C11H14O5, 226. Anal. Calcd for C11H14O5: C, 58.40;
H, 6.24. Found: C, 58.52; H, 6.43. λmax(ꢀ) ) 358 (1890) nm.
2-(2′-(2′′-(2′′′-Methoxyethoxy)ethoxy)ethoxy)-quinone (1c). So-
dium hydroxide (2.4 g, 60 mmol) was added to a stirred solution of
2-benzyloxy-phenol, 5 (3.0 g, 15 mmol), and triethyleneglycol monom-
ethyl ether tosylate, 6c (4.8 g, 15 mmol), in dry dioxane (150 mL),
and then the solution was heated at 80 °C for 24 h. The precipitate
was filtered off, and the filtrate was evaporated under the reduced
pressure to give a viscous residue. The residue was extracted with
benzene to give 2-(2′-(2′′-(2′′′-methoxyethoxy)ethoxy)ethoxy)-1-ben-
zyloxybenzene, 7c (4.9 g, 94%). Subsequently, a mixture of 6c and Pd
carbon (1 g) in dioxane (10 mL) was stirred under an atmosphere of
hydrogen for 3 days. The catalyst was filtered off, and the solvent was
evaporated under the reduced pressure to give 2-(2′-(2′′-(2′′′-methoxy-
ethoxy)ethoxy)ethoxy)phenol, 8c (3.27 g, 89%), as an oil. Fremy’s salt
(3.8 g, 14 mmol) was added to a stirred suspension of 8c (0.81 g, 3
mmol) in 5% aqueous sodium acetate (280 mL) containing a small
amount of benzene (30 mL). After the mixture had been stirred at room
temperature for 30 min, it was extracted with benzene (3 × 70 mL).
The combined organic layers were dried (MgSO4) and concentrated
under reduced pressure. The residue was chromatographed on a silica
gel with benzene-chloroform (1:3). After removal of solvent, 1c was
obtained as a yellow oil (0.42 g, 52%). 1c: 1H NMR (270 MHz, CDCl3)
δ 3.38 (s, 3H, Me), 3.53-3.57 (m, 2H, oxymethylene), 3.62-3.68 (m,
4H, oxymethylene), 3.70-3.74 (m, 2H, oxymethylene), 3.88-3.92 (m,
2H, oxymethylene), 4.07-4.10 (m, 2H, oxymethylene), 5.96 (s, 1H,
quinone), 6.69 (s, 1H, quinone), and 6.70 (s, 1H, quinone); IR (KBr)
2881, 1683, 1654, 1591, 1350, 1308, 1232, 1193, 1099, 942, and 871
cm-1; FAB-MS m/z 271 (M + 1), calcd for C13H18O6, 270. Anal. Calcd
for C13H18O6: C, 57.77; H, 6.71. Found: C, 57.49; H, 6.65. λmax(ꢀ) )
359 (1820) nm.
2,3-Bis(2′-(2′′-methoxyethoxy)ethoxy)-quinone (2b). Sodium hy-
droxide (8.0 g, 200 mmol) was added to a stirred solution of
3-benzyloxypyrocatechol, 9 (5.0 g, 23 mmol), and diethyleneglycol
monomethyl ether tosylate, 6b (12.6 g, 46 mmol), in dry dioxane (200
mL), and then the solution was heated at 80 °C for 24 h. The precipitate
was filtered off, and the filtrate was evaporated under the reduced
pressure to give a viscous residue. The residue was extracted with
benzene to give 2,3-bis(2′-(2′′-methoxyethoxy)ethoxy)-1-benzyloxy-
benzene, 10b (8.6 g, 89%). Subsequently, a mixture of 10b and Pd
carbon (1 g) in dioxane (10 mL) was stirred under an atmosphere of
hydrogen for 3 days. The catalyst was filtered off, and the solvent was
evaporated under the reduced pressure to give 2,3-bis(2′-(2′′-methoxy-
ethoxy)ethoxy)phenol, 11b (6.3 g, 93%), as an oil. Fremy’s salt (3.8
g, 14 mmol) was added to a stirred suspension of 11b (1.0 g, 3 mmol)
in 5% aqueous sodium acetate (280 mL) containing a small amount of
benzene (30 mL). After the mixture had been stirred at room
temperature for 30 min, it was extracted with benzene (3 × 70 mL).
The combined organic layers were dried (MgSO4) and concentrated
under reduced pressure. The residue was chromatographed on a silica
gel with chloroform-THF (29:1). After removal of solvent, 2b was
obtained as a red oil (0.38 g, 38%). 2b: 1H NMR (270 MHz, CDCl3)
δ 3.36 (s, 6H, Me), 3.50-3.53 (m, 4H, oxymethylene), 3.63-3.67 (m,
4H, oxymethylene), 3.76-3.78 (m, 4H, oxymethylene), 4.42-4.45 (m,
4H, oxymethylene), and 6.58 (s, 2H, quinone); IR (KBr) 2924, 1656,
1590, 1459, 1295, 1109, 1074, and 843 cm-1; FAB-MS m/z 345 (M +
1), calcd for C16H24O8, 344. Anal. Calcd for C16H24O8: C, 55.81; H,
7.02. Found: C, 55.79; H, 7.11. UV/vis (CHCl3): λmax(ꢀ) ) 271 (2880)
and 398 (1010) nm.
2,3-Bis(2′-(2′′-(2′′′-methoxyethoxy)ethoxy)ethoxy)-quinone (2c).
Sodium hydroxide (8.0 g, 200 mmol) was added to a stirred solution
of 3-benzyloxypyrocatechol, 9 (5.0 g, 23 mmol), and triethyleneglycol
monomethyl ether tosylate, 6c (14.6 g, 46 mmol), in dry dioxane (200
mL), and then the solution was heated at 80 °C for 24 h. The precipitate
was filtered off, and the filtrate was evaporated under the reduced
pressure to give a viscous residue. The residue was extracted with
benzene to give 2,3-bis(2′-(2′′-(2′′′-methoxyethoxy)ethoxy)ethoxy)-1-
benzyloxybenzene, 10c (10.3 g, 88%). Subsequently, a mixture of 10c
and Pd carbon (1 g) in dioxane (10 mL) was stirred under an atmosphere
of hydrogen for 3 days. The catalyst was filtered off, and the solvent
was evaporated under the reduced pressure to give 2,3-bis(2′-(2′′-(2′′′-
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5821

A R T I C L E S
Tsuda et al.
methoxyethoxy)ethoxy)ethoxy)phenol, 11c (7.9 g, 89%), as an oil.
Fremy’s salt (3.8 g, 14 mmol) was added to a stirred suspension of
11c (1.3 g, 3 mmol) in 5% aqueous sodium acetate (280 mL) containing
a small amount of benzene (30 mL). After the mixture had been stirred
at room temperature for 30 min, it was extracted with benzene (3 ×
70 mL). The combined organic layers were dried (MgSO4) and
concentrated under reduced pressure. The residue was chromatographed
on a silica gel with chloroform-THF (14:1). After removal of solvent,
2c was obtained as a red oil (0.46 g, 37%). 2c: 1H NMR (270 MHz)
CDCl3 δ 3.37 (s, 6H, Me), 3.51-3.55 (m, 4H, oxymethylene), 3.60-
3.69 (m, 12H, oxymethylene), 3.75-3.79 (m, 4H, oxymethylene),
4.41-4.44 (m, 4H, oxymethylene), and 6.58 (s, 2H, quinone); IR (KBr)
2878, 1656, 1589, 1459, 1294, 1106, and 845 cm-1; FAB-MS m/z 433
(M + 1), calcd for C20H32O10, 432. Anal. Calcd for C20H32O10: C, 55.55;
H, 7.46. Found: C, 55.29; H, 7.66. UV/vis (CHCl3): λmax(ꢀ) ) 271
(2460) and 400 (1020) nm.
Supporting Information Available: NMR spectra (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
JA021444T
9
5822 J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003