Evidence for complexes of different stoichiometries between organic solvents and cyclodextrins

Article abstract of DOI:10.1039/b513214b

The influence of the organic solvent on the acid and basic hydrolysis of N-methyl-N-nitroso-p-toluenesulfonamide (MNTS) in the presence of α- and β-cyclodextrins has been studied. The observed rate constant was found to decrease through the formation of an unreactive complex between MNTS and the cyclodextrins. In the presence of dioxane, acetonitrile or DMSO, the inhibitory effect of β-CD decreased on increasing the proportion of organic cosolvent as a result of a competitive reaction involving the formation of an inclusion complex between β-CD and the cosolvent. The disparate size of the organic solvent molecules resulted in stoichiometric differences between the complexes; the β-CD-dioxane and β-CD-DMSO complexes were 1: 1 whereas the β-CD-acetonitrile complex was 1: 2. The basic and acid hydrolysis of MNTS in the presence of α-CD showed a different behavior; thus, the reaction gave both 1: 1 and 2: 1 α-CD-MNTS complexes, of which only the former was reactive. This result was due to the smaller cavity size of α-CD and the consequent decreased penetration of MNTS into the cavity in comparison to β-CD. The acid hydrolysis of MNTS in the presence of α-CD also revealed decreased penetration of MNTS into the cyclodextrin cavity, as evidenced by the bound substrate undergoing acid hydrolysis. In addition, the acid hydrolysis of MNTS in the presence of acetonitrile containing α-CD gave 1: 1 α-CD-acetonitrile inclusion complexes, which is consistent with a both a reduced cavity size and previously reported data. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b513214b

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Evidence for complexes of different stoichiometries between organic solvents
and cyclodextrins†
L. Garc´ıa-R´ıo,*a P. Herve´s,b J. R. Leis,a J. C. Mejuto,c J. Pe´rez-Justeb and P. Rodr´ıguez-Dafontea
Received 19th September 2005, Accepted 1st January 2006
First published as an Advance Article on the web 30th January 2006
DOI: 10.1039/b513214b
The influence of the organic solvent on the acid and basic hydrolysis of N-methyl-N-nitroso-p-
toluenesulfonamide (MNTS) in the presence of a- and b-cyclodextrins has been studied. The observed
rate constant was found to decrease through the formation of an unreactive complex between MNTS
and the cyclodextrins. In the presence of dioxane, acetonitrile or DMSO, the inhibitory effect of b-CD
decreased on increasing the proportion of organic cosolvent as a result of a competitive reaction
involving the formation of an inclusion complex between b-CD and the cosolvent. The disparate size of
the organic solvent molecules resulted in stoichiometric differences between the complexes; the
b-CD–dioxane and b-CD–DMSO complexes were 1 : 1 whereas the b-CD–acetonitrile complex was 1 :
2. The basic and acid hydrolysis of MNTS in the presence of a-CD showed a different behavior; thus,
the reaction gave both 1 : 1 and 2 : 1 a-CD–MNTS complexes, of which only the former was reactive.
This result was due to the smaller cavity size of a-CD and the consequent decreased penetration of
MNTS into the cavity in comparison to b-CD. The acid hydrolysis of MNTS in the presence of a-CD
also revealed decreased penetration of MNTS into the cyclodextrin cavity, as evidenced by the bound
substrate undergoing acid hydrolysis. In addition, the acid hydrolysis of MNTS in the presence of
acetonitrile containing a-CD gave 1 : 1 a-CD–acetonitrile inclusion complexes, which is consistent with
a both a reduced cavity size and previously reported data.
It is clear that most studies concerning the stability of cyclodex-
trin complexes have been conducted in pure aqueous solutions;
however, a number of authors have examined the formation
Introduction
Cyclodextrins are cyclic oligosaccharides1 and a-, b- and c-
cyclodextrins consist of 6, 7 and 8 a-amylose units, respectively.
of cyclodextrin complexes in aqueous–organic binary solvent
These macrocycles can act as hosts to form inclusion complexes
mixtures or in pure organic solvents. There is evidence that the
with guest molecules in the solid state or in solution. The shape of
nature of the solvent can influence or even determine the structure
of the complex.4 Indeed, some cyclodextrin complexes have been
found to be more stable in mixed solvents and pure organic
media than in aqueous solutions5 or more stable in D2O than
H2O.6 However, the solvent effect usually reduces the stability
of the complex relative to water.7 Most studies in this context
have used aqueous mixtures with common water-miscible solvents
such as alcohols, dimethyl sulfoxide (DMSO), dimethylformamide
(DMF), dioxane or acetonitrile; however, complexes have also
been found to form in pure solvents of these types8 as well as in
hydrocarbons and similar nonpolar solvents.5d,9
a cyclodextrin molecule is a hollow, truncated cone (a toroid) of
2
˚
height 7.8 A, which represents the width of the amylose unit. The
diameter of the cavity depends on the number of amylose units
present.3 All hydroxyl groups in a cyclodextrin are projected
outward, so the cavity is hydrophobic and nonpolar. Thus,
despite the prevalence of hydroxyl groups, cyclodextrins possess
a hydrophobic surface to which organic molecules in aqueous
solutions can bind hydrophobically. The primary requirement for
a guest to be included in a cyclodextrin cavity is obviously that its
size should be smaller than the cavity itself. If the guest molecule
is of an appropriate size, then the hydrogen bonding network can
be extended within the cavity through the open ends of the toroid
and a hydration sheath can be formed around the guest molecule.
A supramolecule containing a cyclodextrin molecule as the host
is perhaps the model that best describes an enzyme–substrate
complex.
The effects of solvents on the stability of cyclodextrin com-
plexes have been examined in light of various hypotheses.10 One
approach is based on the assumption that organic cosolvents
act as competitive substrates; as their concentration is raised,
their ability to compete in displacing the primary substrate
from the guest cavity increases.11 This hypothesis allows one
to evaluate binding constants for the postulated cosolvent–
cyclodextrin complex formation reactions, the results of which
are usually chemically reasonable. Another hypothesis involves
the assumption that the organic cosolvent undergoes inclusion
alongside the primary substrate to form a complex in a 1 : 1 : 1—
or higher—proportion.5a,5b,5e,12 This hypothesis accounts for the
increased complex stability observed when an organic cosolvent is
added to an aqueous system.
aDepartamento de Qu´ımica F´ısica, Facultad de Qu´ımica, Universidad de
Santiago, 15782, Santiago, Spain
bDepartamento de Qu´ımica F´ısica, Facultad de Ciencias, Universidad de
Vigo, Vigo, Spain
cDepartamento de Qu´ımica F´ısica, Facultad de Ciencias, Universidad de
Vigo, Ourense, Spain
† Electronic supplementary information (ESI) available: Kinetic data,
derivation of equations. See DOI: 10.1039/b513214b
1038 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is
The Royal Society of Chemistry 2006
©

View Online
In the work described here, we used a kinetic method to study
the complexation of dioxane, DMSO and acetonitrile by a- and
b-cyclodextrin. We examined the influence of the dioxane, DMSO
and acetonitrile concentrations on the acid and basic hydrolysis of
N-methyl-N-nitroso-p-toluenesulfonamide (MNTS) (Scheme 1)
in the presence of a- and b-CD. Both reactions were previously
examined in an aqueous medium.13
order conditions ([MNTS] = 1.56 × 10−5 M ꢀ [OH−] or [H+]).
The integrated first-order rate law was fitted to the absorbance–
time data by linear regression (r > 0.999) in all cases. The
observed rate constants, kobs, were thus reproduced to within
3%. In the tests that did not involve organic solvent, MNTS
was dissolved in water containing CD and a small volume of the
resulting solution was subsequently added to the reaction mixture;
in experiments involving an organic cosolvent, MNTS was first
dissolved in dioxane, DMSO or acetonitrile and further solvent
was subsequently added to the reaction medium in the required
proportion.
Results and discussion
In this section the results obtained in the acid and basic hydrolysis
of MNTS in the presence of a- or b-CD in dioxane, acetonitrile or
DMSO as organic cosolvent are presented and discussed.
Scheme 1
1. b-Cyclodextrin
The results allowed us to determine the binding constants
for the complexes formed between the organic solvents and the
cyclodextrins and also gave the stoichiometries of the complexes.
The most salient finding is that, while dioxane and DMSO form
1 : 1 complexes with a- and b-CD, acetonitrile forms host–guest
complexes with a 1 : 1 stoichiometry with a-CD and a 1 : 2
stoichiometry with b-CD. It should be noted that these organic
molecules are frequently used as solvating agents in studies on the
catalytic efficiency of cyclodextrins.
1a. Acid and basic hydrolysis of MNTS in the presence of
dioxane and DMSO. The influence of the b-CD concentration
on kobs for the acid hydrolysis of MNTS in dioxane over a range
of concentrations as an organic cosolvent is illustrated in Fig. 1a.
As can be seen, kobs decreased on increasing the concentration of
b-CD. These results are consistent with a mechanism involving
the formation of an unreactive complex between MNTS and the
cyclodextrin. In the absence of dioxane, the observed rate constant
decreased by a factor of about 10 on increasing the cyclodextrin
concentration from 4.00 × 10−4 M to 8.00 × 10−3 M. In the
presence of dioxane, the inhibitory effect decreased from about
7 to 2.3 times on increasing the cosolvent concentration from
1.95 × 10−2 to 0.234 M.
Experimental
MNTS, spectroscopic grade solvents (dioxane, DMSO and ace-
tonitrile) and the cyclodextrins were purchased from Sigma or
Fluka and used as received. Cyclodextrin solutions were prepared
taking into account the amount of water contained in commercial
a- and b-CD.
The kinetic procedures largely conformed to well-established
practices.13 Reactions were monitored through the first-order acid
or basic hydrolysis of MNTS at 250 nm, using a Hewlett-Packard
8453 spectrophotometer with the observation cell thermostated at
(25.0 0.1) ◦C. All kinetic tests were conducted under pseudo first-
The results outlined above rule out the formation of mixed
inclusion complexes with MNTS and the organic cosolvent. Thus,
ethanol forms inclusion complexes with b-CD and the small size
of this system may allow the co-inclusion of ethanol and the
substrate.14 The inclusion of the organic cosolvent in the b-CD
cavity must increase its hydrophobicity through the displacement
of water molecules from within. The increased hydrophobicity that
results should lead to an increase in the binding constant of MNTS
Fig. 1 (a) Influ◦ence of the concentration of b-CD on kobs for the acid hydrolysis of MNTS in the presence of 0.160 M and variable concentrations of
dioxane at 25.0 C. (᭺) [Dioxane] = 1.95 × 10−2 M; (᭹) [dioxane] = 3.90 × 10−2 M; (ꢀ) [dioxane] = 7.03 × 10−2 M; (ꢁ) [dioxane] = 0.117 M; (ꢂ)
[dioxane] = 0.172 M; and (ꢁ) [dioxane] = 0.234 M. (b) Fitting of the data to eqn (2).
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1038–1048 | 1039
©

View Online
b-CDH
to b-CD as the proportion of organic solvent is raised. However,
our results for dioxane, DMSO and acetonitrile contradict this
assumption and do not support the formation of mixed inclusion
complexes in this situation.
an aqueous medium, K
the equilibrium binding constant of
MNTS
b-CDH
MNTS to unionized cyclodextrin and K
of the inclusion complex between dioxane and the neutral form
of the cyclodextrin. From Scheme 2 it is possible to derive the
that for the formation
dioxane
following expressions:17
For a long time it was widely believed that cyclodextrins formed
inclusion complexes only in pure aqueous solutions, mainly
because attempts to induce precipitation of CD adducts from
organic solvents failed. The first successful study concerning the
influence of the solvent on the formation of inclusion complexes
with CD was described by Siegel and Breslow,15 who determined
the stability constants for the complexes of various organic species
with b-CD in DMF, DMSO and water–DMSO mixtures. Their
results led to the following conclusions: (1) the solvent plays a
key role in the binding process; (2) the complexes formed in
nonaqueous and water–organic mixed solvents are weaker than
those obtained in pure aqueous solutions. However, Nelson et al.16
determined the stability constants of the complexes formed by
pyrene with b- and c-CD in the presence of 10% tert-butyl alcohol
as being one order of magnitude greater than those obtained in
pure aqueous solutions. Furthermore, other studies5e have shown
b-CD and hydroxypropyl-b-CD to form ternary complexes in the
presence of alcohols.
The effect of cyclodextrins on the acid hydrolysis of MNTS is
due to the formation of an inclusion complex between MNTS
and the cyclodextrin. Our kinetic results in the presence of a
cosolvent are consistent with a competitive process involving the
binding of dioxane molecules to the cyclodextrin. Raising the
dioxane concentration increased the amount of b-CD that binds
the organic cosolvent and, as a result, decreased the quantity of b-
CD available for binding MNTS (see Scheme 2). As a consequence,
the inhibitory effect of b-CD on the acid hydrolysis of MNTS
decreases on increasing the concentration of organic solvent.
b-CDH
kw[H+]
1 + KMNTS
b-CDH[CD]free
K
MNTS
kobs
=
Kapp
=
(1)
b-CDH
1 + K
[dioxane]
dioxane
T
The pseudo first-order rate constant can be written in linear form
as:
1
1
Kapp
=
+
[CD]T
(2)
kobs
kw [H+] kw [H+]
The good fit between the experimental results obtained at variable
dioxane concentrations and eqn (2) is shown in Fig. 1b. The Kapp
and kw values obtained at each organic cosolvent concentration
studied are shown in Table 1. It can be seen that both constants
decrease on increasing the proportion of dioxane in the reaction
medium. The variation of kw with the dioxane concentration
is consistent with the influence of the solvent polarity on the
reaction rate in the absence of cyclodextrin.18 On the other hand,
the variation of Kapp with the proportion of organic cosolvent
conforms to the following equation:
1
1
Kb-CDH
=
+
dioxane [dioxane]T
(3)
b-CDH
MNTS
b-CDH
Kapp
K
K
MNTS
As can be seen from Fig. 2a, the acid hydrolysis of MNTS in
the presence of b-CD and dioxane fits eqn (3) quite closely. This
equation allowed us to determine the binding constant of MNTS
to unionized cyclodextrin, K
b-CDH
MNTS
= (1380
220) M−1, which
is consistent with the value obtained in the absence of organic
b-CDH
MNTS
cosolvent: K
= (1530 90) M−1. Similarly, we were able to
obtain the binding constant of dioxane to the unionized form of
= (23.8 3.7) M−1 (see Tables 2 and 3).
b-CDH
b-CD− K
dioxane
The experimental results for the basic hydrolysis of MNTS in the
presence of variable concentrations of dioxane exhibited a similar
trend to those for the acid hydrolysis (refer to the ESI). Thus, kobs
decreased on increasing the proportion of CD in the reaction
medium. This was a result of the formation of an unreactive
inclusion complex between MNTS and the ionized form of the
−
cyclodextrin, with KMb-NCDTS . As the proportion of dioxane in the
reaction medium was raised, a competitive equilibrium between
the complexation of MNTS and dioxane by the anionic form
of cyclodextrin was established. Consequently, the experimental
results can be interpreted in terms of a mechanism similar to that
in Scheme 2 and with rate equations similar to (2) and (3). This
Scheme 2
In the reaction mechanism depicted in Scheme 2, kw is the
bimolecular rate constant for the acid hydrolysis of MNTS in
Table 1 Kinetic constants obtained by fitting eqn (2) to the experimental results for the acid and basic hydrolysis of MNTS in the presence of b-CD and
variable concentrations of dioxane at 25.0 ◦
C
[NaOH] = 0.175 M
[HCl] = 0.160 M
b-CD−
[Dioxane]/M
K
/M−1
102 kw/M−1 s−1
[Dioxane]/M
Kabp-pCDH/M−1
102 kw/M−1 s−1
app
0
1690 280
886 39
650 20
539 14
3.21 0.54
2.83 0.12
2.89 0.09
2.95 0.08
2.99 0.03
2.81 0.03
2.85 0.04
0.0156
0.039
0.117
0.234
0.586
0.937
275 11
7.51 0.30
9.24 0.15
9.27 0.16
9.19 0.18
9.65 0.29
9.59 0.59
0.0195
0.039
0.0703
0.117
0.172
0.232
212
103
64
3
2
1
1
1
386
250
219
4
3
3
30
7
1040 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is
The Royal Society of Chemistry 2006
©

View Online
Fig. 2 (᭺) Variation of Kapp for the acid (a) and basic (b) hydrolysis of MNTS in the presence of b-CD and dioxane as a function of the organic solvent
concentration. (᭹) Fitting of the data to eqn (3).
binding model allowed us to determine Kapp and kw at different
dioxane concentrations (see Table 1). As predicted by eqn (3), both
parameters decrease on increasing the proportion of dioxane. The
good fitting of eqn (3) (in Fig. 2b) allowed us to determine the
binding constants of MNTS and dioxane to the anionic form of
on increasing the concentration of organic cosolvent. Providing
a quantitative explanation for the experimental results entails
considering the twofold effect of the addition of dioxane to a b-CD
solution, namely: (a) the binding of dioxane to the cyclodextrin
reduces the concentration of free b-CD; and (b) because the
concentration of dioxane will normally be much higher than that
of b-CD, the properties of the aqueous medium may be altered by
the presence of the organic cosolvent.
−
−
b-CD
b-CD
cyclodextrin, viz. K
= (325 30) M−1 and K
= (16.8
MNTS
dioxane
1.6) M−1, respectively (see Table 3).
The difference between the binding constant for the anionic
and neutral forms of cyclodextrin is consistent with previously
reported results: the binding constant of the substrate to the
cyclodextrin anion is lower than that to the neutral cyclodextrin.19
The evaluation of [b-CD]free entails solving the following equa-
tion (refer to the ESI for details):
([b-CD]free)2 + [b-CD]free (1 + Kdbi-oCxDanHe [dioxane]T − Kdbi-oCxDanHe
Quantitative analysis of the solvent effect. Our results show that
the addition of dioxane to a b-CD solution leads to the formation
of a 1 : 1 complex between the solvent and the cyclodextrin. This
reduces the amount of b-CD available for binding to MNTS and
results in the inhibitory effect of the addition of CD decreasing
[b-CD]total) − [b-CD]T = 0
(4)
A simulated value for the formation constant of the inclusion
b-CDH
dioxane
complex between dioxane and cyclodextrin, K
, was used to
calculate [b-CD]free at each [b-CD]T value, with the resulting data
Table 2 Kinetic constants obtained by fitting eqn (2) and (3) to the experimental results for the acid and basic hydrolysis of MNTS in the presence of
b-CD and variable concentrations of acetonitrile at 25 ◦
C
[NaOH] = 0.175 M
[HCl] = 0.160 M
b-CD−
[CH3CN]/M
K
/M−1
102 kw/M−1 s−1
[CH3CN]/M
Kabp-pCDH/M−1
102 kw/M−1 s−1
app
0
1690 280
3.21 0.54
3.06 0.02
2.79 0.08
2.77 0.07
2.80 0.08
2.03 0.03
1.83 0.03
0.0957
0.638
0.957
1.910
2.870
3.830
292 10
7.28 0. 26
7.59 0.22
8.24 0.11
8.76 0.12
6.79 0.16
5.46 0.15
0.0638
0.510
0.957
1.468
1.978
2.872
1467
9
252
248
142
53
7
3
2
1
1
1196 33
1009 25
813 24
476
244
7
4
29
Table 3 Thermodynamic parameters for the acid and basic hydrolysis of MNTS in the presence of a- or b-CD and dioxane, DMSO or acetonitrile as
organic cosolvent. The solvent–cyclodextrin and cyclodextrin–MNTS complexes were both 1 : 1
Cyclodextrin
Cosolvent
Eqn
KCMDNHTS/M−1
KMCDNTS/M−1
KCsoDlvHent/M−1
KsCoDlvent/M−1
b-CD
b-CD
None
Dioxane
(1)
(3)
(5)
(3)
(5)
(9)
(3)
(10)
1530 90
1380 220
1180 14
1600 270
1780 50
400 20
325 30
—
—
23.8 3.7
20.0 2.5
2.8 0.5
4.0 0.5
—
9.0 0.6
7.0 1.0
16.8 1.6
20.0 2.5
3.1 0.4
2.5 0.5
—
—
—
363
4
b-CD
DMSO
480 200
430 13
—
—
—
a-CD
a-CD
None
Acetronitrile
47
45
51
4
3
2
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1038–1048 | 1041
©

View Online
Fig. 3 (a) (᭺) Influence of the concentration of free b-CD on kobs/kw for the acid hydrolysis of MNTS in the presence of variable concentrations of
dioxane (data from Fig. 1). (᭹) Fitting of the data to eqn (5). (b) Analysis of the experimental results for the basic hydrolysis of MNTS (data from the
ESI).
b-CDH
employed to solve the rate equation
reaction. The values found were K
= (1600
270) M−1
MNTS
−
b-CD
and K
= (480
200) M−1, respectively. These values are
MNTS
b-CDH
kw
kobs
1
[H+]
K
MNTS
similar to those obtained in water containing dioxane (vide
=
+
[b-CD]free
(5)
−
[H+]
b-CDH
DMSO
b-CD
supra). We also obtained K
= (2.8 0.5) M−1 and K
=
DMSO
(3.1 0.4) M−1.
b-CDH
dioxane
The simulated K
values that best fit eqn (5) were adopted.
The quantitative analysis of the solvent effect on the acid and
basic hydrolysis of MNTS in the presence of b-CD was based
on the same mathematical treatment previously used for diox-
ane. The optimization procedure provided the following values:
The linearity criterion used was the correlation coefficient and
b-CDH
dioxane
the K
value resulting in the highest coefficient in each case
was adopted. The variation of the correlation coefficient used to
b-CDH
dioxane
fit eqn (5) to the experimental data as a function of K
and
b-CDH
b-CDH
MNTS
−
K
= (4.0 0.5) M−1 and K
= (1780 50) M−1 for the
b-CD
DMSO
K
for the acid and basic hydrolysis of MNTS, respectively,
−
−
dioxane
b-CD
acid hydrolysis, and K
= (2.5 0.5) M−1 and KMb-NCDTS = (430
b-CDH
dioxane
DMSO
is given in the ESI. The values obtained, viz. K
= (20.0
15) M−1 for the basic hydrolysis of MNTS, respectively.
−
b-CD
2.5) M−1 and K
= (20.0 2.5) M−1, are consistent with the
dioxane
previous ones. The good fitting of eqn (5) to the results for the
acid (Fig. 3a) and basic (Fig. 3b) hydrolysis of MNTS is shown in
Fig. 3. It should be noted that Fig. 3 includes the results obtained
at variable dioxane concentrations. As can be seen, all data fitted
a single equation and the solvent effect can therefore be assumed
to be twofold, namely: (a) an alteration of the properties of the
medium that can be corrected using the kobs/kw ratio; and (b) a
competitive binding equilibrium involving the organic solvent and
b-CD.
The kinetic implications of the changes in the properties of
the medium are significant; in fact, the acid and basic hydrolysis
of MNTS are both sensitive to the amount of organic cosolvent
present in the reaction medium.13 The term corresponding to the
competitive binding of dioxane arises from the solvent molecule
being sufficiently large to fill most of the b-CD cavity; this excludes
the possibility of a single cyclodextrin molecule simultaneously
binding a dioxane molecule and an MNTS molecule.
The investigation into the influence of organic cosolvents on
the acid and basic hydrolysis of MNTS in the presence of b-
CD was extended to DMSO. The rate constants for the two
reactions were found to decrease on increasing the concentration
of cyclodextrin (see the ESI). As in the previous case, one can
assume the formation of 1 : 1 MNTS–b-CD and DMSO–b-CD
inclusion complexes. The reaction mechanism must be similar to
1b. Acid and basic hydrolysis of MNTS in the presence of
acetonitrile. We also examined the influence of the b-CD con-
centration on the acid and basic hydrolysis of MNTS in media
containing variable concentrations of acetonitrile. The addition of
b-CD was found to decrease kobs in all cases. The inhibition due to
the addition of cyclodextrin becomes less marked on increasing the
proportion of acetonitrile in the reaction medium (see Fig. 4 for the
acid hydrolysis). The addition of high proportions of acetonitrile
caused the kobs value at a zero concentration of b-CD to decrease
on increasing the content of organic cosolvent. This observation is
consistent with the effect of organic solvents on the rate of acid and
basic hydrolysis of MNTS (see the results for the basic hydrolysis
in the ESI).
It can be seen from Fig. 4 that eqn (3) fits the experimental
b-CDH
app
b-CD
app
results quite well. The K
and K
values obtained are
given in Table 2 along with the rate constants for the reactions
in the absence of cyclodextrin and the presence of variable
concentrations of acetonitrile. In contrast to the previous results—
and the predictions of eqn (3)—a linear relationship between
b-CDH
app
b-CDH
app
1/K
or 1/K
and the acetonitrile concentration was
not observed; rather, the plot (not shown) was clearly curved,
which suggests that the stoichiometry of the complex between
acetonitrile and b-CD is higher than 1 : 1. The formation of a
2 : 1 acetonitrile–b-CD complex is illustrated in Scheme 3.
Bearing in mind that the concentration of cyclodextrin bound
to MNTS was negligible relative to the combination of free
and acetonitrile-bound cyclodextrin, if one assumes that the
concentration of acetonitrile bound to cyclodextrin is negligible
that shown in Scheme 2 and the resulting equations similar to
b-CDH
those previously formulated for dioxane, with K
instead of
DMSO
b-CDH
K
as the equilibrium constant. The use of eqn (3) enabled us
dioxane
to determine the binding constant of MNTS to the ionized and
unionized forms of cyclodextrin in the acid and basic hydrolysis
1042 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is
The Royal Society of Chemistry 2006
©

View Online
Fig. 4 (a) Influence of the concentration of b-CD on kobs for the acid hydrolysis of MNTS in the presence of variable concentrations of acetonitrile. (᭺)
[CH3CN] = 6.38 × 10−2 M; (᭹) [CH3CN] = 0.510 M; (ꢀ) [CH3CN] = 0.957 M; (ꢁ) [CH3CN] = 1.468 M; (ꢁ) [CH3CN] = 1.978 M and (ꢂ) [CH3CN] =
2.872 M. [HCl] = 0.160 M; T = 25.0 ◦C. (b) Plot of 1/kobs as a function of [b-CD] based on eqn (2).
to determine the binding constants of MNTS to b-CD in acidic and
b-CDH
basic media. These constants were found to be K
= (1690
MNTS
−
200) M−1 and KMb-NCDTS = (495 240) M−1, respectively. These values
are consistent with the results previously obtained in the absence
of an organic solvent. The relationship between the slope and
the intercept allowed us to determine the equilibrium formation
constants for the (CH3CN)2–b-CD complex in acidic and basic
−
media: Kb-CDCHN = (0.70 0.08) M−1 and KCb-HCDCN = (1.05 0.50) M−1,
CH
3
3
respectively.
Scheme 3
Quantitative analysis of the solvent effect. Once the stoichiome-
try of the inclusion complex formed by acetonitrile and b-CD had
been determined, we sought to confirm whether the proposed re-
action scheme would account for the body of experimental results.
The proposed model assumes that the properties of the aqueous
medium are altered by the addition of the organic cosolvent. The
effect of this addition on the reactivity can be suppressed by using
the kw/kobs ratio, where kw is the rate constant in the absence of
cyclodextrin. We studied aqueous media consisting of either water
alone or water–organic cosolvent mixtures. We also assumed that
the cyclodextrin established a competitive equilibrium where the
binding of acetonitrile molecules decreased the amount of b-CD
available for binding MNTS.
b-CDH
relative to the total concentration of the solvent, then K
be expressed as follows:
can
(6)
app
b-CDH
K
MNTS
b-CDH
K
=
=
;
app
b-CDH
1 + K
[CH3CN]2
T
CH CN
3
Kb-CDH
1
1
+
CH CN [CH3CN]2
3
T
b-CDH
MNTS
b-CDH
CDH
Kapp
K
K
MNTS
Eqn (6) predicts a linear relationship between 1/Kabp-pCDH or 1/K
b-CDH
app
and [CH3CN]2. The results obtained for the acid and basic
hydrolysis of MNTS (Fig. 5) are quite consistent with this
prediction and hence with the presence of a 2 : 1 complex between
acetonitrile and b-CD. Plots such as that shown in Fig. 5 were used
The mass balances for MNTS, acetonitrile and the cyclodextrin
yield the equation
b-CD
b-CD
K
[CH3CN]3 + (2 K
[b-CD]T − Kb-CD
CH CN
3
CH CN
CH CN
3
free
[CH3CN]T)[CH3CN]2free + 3[CH3CN]free − [CH3CN]T = 0 (7)
This third-order equation was solved by using the method
described previously: a simulated value for the formation constant
of the complex between acetonitrile and b-CD was used to
calculate the concentration of unbound acetonitrile [eqn (7)].
The relationship between the observed rate constant and the free
cyclodextrin concentration, eqn (5), was then used to identify the
b-CDH
3
K
CN value that provided the free b-CD concentrations that best
CH
fitted eqn (5). In this way, the following values were obtained:
b-CDH
3
K
= (0.60 0.05) M−2 and KCb-HCDH = (0.50 0.05) M−2.
CH CN
CN
3
The results obtained in this way for the acid and basic hydrolysis
of MNTS are shown in Table 4 and Fig. 6. As can be seen, all
experimental data fitted a single curve—a fact that confirms the
accuracy of the proposed method, which assumes the solvent effect
to encompass two contributions: namely, the alteration of the
b-CDH
Fig. 5 Fitting of eqn (6) to the 1/K
values obtained as a function
app
of [CH3CN]2 in the acid (᭺) and basic (᭹) hydrolysis of MNTS in the
presence of b-CD.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1038–1048 | 1043
©

View Online
Table 4 Thermodynamic parameters for the acid and basic hydrolysis of MNTS in the presence of a- or b-CD. The complex formed by b-CD with
acetonitrile as the cosolvent was 2 : 1 (see Scheme 3) and that formed by a-CD with MNTS in the basic hydrolysis reaction was 1 : 2 (see Scheme 4)
Cyclodextrin
b-CD
Cosolvent
Acetonitrile
None
Eqn
KCMDNHTS/M−1
KMCDNTS/M−1
KCsoDlvHent/M−2
KsCoDlvent/M−2
(6)
(5)
(8)
1690 200
1490 15
—
495 240
0.70 0.09
0.60 0.05
—
1.05 0.50
0.50 0.05
—
340
4
a-CD
2.6 0.3
Fig. 6 (a) (᭺) Influence of the concentration of free b-CD on kobs/kw for the acid hydrolysis of MNTS in the presence of variable concentrations of
acetonitrile (data from Fig. 4). (᭹) Fitting of eqn (5) to the data. (b) Analysis of the experimental results for the basic hydrolysis of MNTS (data from the
ESI).
properties of water and the competitive binding of the organic
solvent to the cyclodextrin.
2. a-Cyclodextrin
2A. Basic hydrolysis of MNTS. The basic hydrolysis of
MNTS in the presence of b-CD is strongly inhibited by the
formation of an unreactive complex between MNTS and the
cyclodextrin. The presence of a-CD results in a distinct type
of behavior: kobs increases moderately to a maximum value on
increasing the a-CD concentration but beyond this maximum the
constant decreases on further increasing [a-CD] (see Fig. 7). The
experimental results can be interpreted in terms of the mechanism
shown in Scheme 4.
Our results contradict those of Taraszewska,20 who studied
the binding of ethanol, DMSO, DMF, acetone and acetonitrile
to b-CD and obtained constants similar to those found for
a-CD and CD–solvent complexes of 1 : 1 stoichiometry, as
b-CD
1 : 1
well as a binding constant of K
= 6.0 M−1 for the b-
CD–acetonitrile complex. It should be noted that the binding
constants varied with the proportion of organic solvent. This is
because the change in the concentration of free cyclodextrin on
increasing the amount of the organic solvent in the mixture was
ignored. Furthermore, Taraszewska found differences between the
complexes of acetonitrile and acetone and those of the other
solvents, and assumed that the zone inside the b-cavity where
o-chloronitrobenzene could be accommodated in the presence
of acetone and acetonitrile was more hydrophobic than in the
presence of other solvents. Park21 studied the complexes of
various organic solvents, including acetonitrile, with b-CD by
spectrophotometry. In that study the inhibitory effect of solutes
on the binding of CD by phenolphthalein was investigated. All
complexes were 1 : 1 and the resulting binding constant of
b-CD
1 : 1
acetonitrile to b-CD was K
= 0.54 M−1. Park’s values for the
Scheme 4
binding constant of acetonitrile to b-CD are consistent with our
own; however, the stoichiometry of the complex is different. We
believe that Park’s results may have been influenced by the fact that
the acetonitrile concentrations considered only covered a small
range (viz. 0.73–0.86 M). In the study described here, acetonitrile
concentrations from ca. 6.00 × 10−2 M to 3.00 M were considered.
The wider range used here may have allowed us to detect the
presence of 2 : 1 complexes. Other authors have also studied the
displacement of phenolphthalein from its complex with b-CD by
the addition of tetrahydrofuran.22
Such a mechanism enables the following rate law to be derived:
−
kw[OH−] + kCDKa-CD 1 : 1[a-CD]T
MNTS
kobs
=
(8)
−
−
-
a-CD
1 : 1
1 + K
1 : 1[a-CD]T + KMa-NCDTS
K
a-CD 1 : 2[a-CD]2T
MNTS
MNTS
where kw and kCD are the rate constants for hydrolysis by OH−
ions in water and by the alkoxide groups in the cyclodextrin,
−
−
a-CD
MNTS
a-CD
MNTS
respectively; and K
1 : 1 and K
1 : 2 are the equilibrium binding
constants for the 1 : 1 complex and 1 : 2 complex [MNTS-(CD)2]
1044 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is The Royal Society of Chemistry 2006
©

View Online
MNTS. Such a reaction was studied in our laboratory23 and
its rate was found to exhibit a Brønsted-type relationship with
the basicity of the starting alcohol for methanol, isopropanol,
tetrafluoroethanol and hexafluoroisopropanol, with a slope bnucl
=
0.49 0.01. The reactivity of the OH− ion was lower than expected
from its basicity, probably due to the need for desolvation prior
to nucleophilic attack, as in other hydroxyl ion- and thiolate-
catalyzed reactions.24 The observed rate constant for the reaction
via the ionized hydroxyl group in the cyclodextrin was roughly half
that expected for an alkoxide with a similar pKa (e.g. CF3CH2O−).
The results are consistent with the formation of 1 : 1 and 1 : 2
MNTS–a-CD complexes, as well as with transition states that
involve a cyclodextrin molecule. These differences between the
behavior of a-CD and b-CD must be a result of geometric factors.
The binding of guests to cyclodextrins is determined largely by
two factors: namely, the size of the included moiety and its
hydrophobicity.25 The disparate binding abilities of a- and b-CD
Fig. 7 Influence of the concentration of a-CD on kobs for the basic
hydrolysis of MNTS. [OH−] = 0.162 M, T = 25.0 ◦C.
between MNTS and the anionic form of a-CD, respectively.
By fitting eqn (8) to the experimental data, and keeping the
kw value constant (kw = 0.083 M−1 s−1), we obtained kCD = (3.1
˚
must be ascribed to the widths of their cavities (4.5 and 7.0 A,
˚
respectively) since their depths are identical (7.0 A). The energy-
−
−
a-CD 1 : 1
a-CD 1 : 2
0.1) × 10−2 s−1, K
= (15 2) M−1 and K
= (2.6
minimized structures obtained using the MNTS dimensions and
optimal geometry derived from molecular mechanics calculations
are shown in Scheme 5. Based on the size of MNTS, the molecule
can be fully accommodated inside the b-CD cavity (1), so the
sulfonyl group will hardly be able to approach the ionized RO−
group in b-CD.
MNTS
MNTS
0.3) M−2.
Although kCD (in s−1) and kw (in M−1 s−1) cannot be compared
directly, the fact that the latter exceeded the former is worthy
of note. The hydrolysis reaction via the MNTS–CD complex
is equivalent to nucleophilic attack of an alkoxide ion on
Scheme 5
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1038–1048 | 1045
©

View Online
In the presence of a-CD, the toluene group in MNTS is too large
to be fully accommodated inside the a-CD cavity. Therefore, the
binding constant of MNTS to both the neutral form (vide infra)
and the anionic form of a-CD will be quite small relative to that for
b-CD. The kinetics of the basic hydrolysis of MNTS with ionized
a-CD clearly indicate that both 2 : 1 and 1 : 1 binding situations
are substantial. The initial increase in the rate constant suggests
that the former binding stoichiometry involves the toluene group
(2), whereas the latter involves the nitroso group (3).
rate constant can be expressed as
kw[H+] + kCHDKaap-pCDH[H+][a-CD]free
kobs
=
(9)
1 + Kaap-pCDH[a-CD]free
a-CDH
app
where K
is the apparent binding constant of MNTS to a-
CD. As can be seen from Fig. 8, this equation closely fits the
results obtained at variable concentrations of acetonitrile. The
Kaap-pCDH, kw and kHCD values obtained for different concentrations of
cosolvent are shown in Table 5. As the acetonitrile concentration
in the reaction medium was increased, the polarity and the rate
constants kw and kHCD decreased. The error in the determination of
kHCD increased with increasing concentration of acetonitrile in the
2B. Acid hydrolysis of MNTS in the presence of acetonitrile.
We studied the influence of a-CD on the acid hydrolysis of
MNTS in the presence of variable amounts of acetonitrile. As an
illustrative example, the results obtained in the presence of variable
concentrations of cosolvent ranging from 3.19 × 10−2 to 0.511 M
are shown in Fig. 8. The resulting kobs values differed markedly
from those found in the presence of b-CD (see Fig. 4). As can be
seen in Fig. 8b, the variation of the rate constant with the a-CD
concentration (dashed line) deviated strongly from linearity.
This finding is inconsistent with both eqn (2) and the reaction
mechanism shown in Scheme 2, but is consistent with the
mechanism in Scheme 6.
reaction medium. This trend is a result of the fact that Kaap-pCDH was
low at high acetonitrile concentrations, where k ꢂ kCHD
K
a-CDH
app
w
[a-CD]free
.
The ability of the a-CD-bound substrate to react with H+ ions
in the aqueous medium is due to the geometry of the inclusion
complex. As shown in 1–3 (Scheme 5), the size of a-CD is sufficient
to accommodate the toluene group in MNTS, but the nitroso
group remains outside—a situation that facilitates attack by H+
ions in the medium. This situation cannot occur in b-CD as its
cavity is large enough to accommodate the nitroso group as well.
It can be seen from Table 5 that raising the acetonitrile
concentration in the reaction medium led to a decrease in
Kaap-pCDH, a change due to the formation of an inclusion complex
between CH3CN and a-CD. Therefore, the variation of Kaap-pCDH
with the acetonitrile concentration should fit an equation similar
to (3). This assumption was confirmed. Fitting eqn (3) to
the experimental results (Fig. 9) provided the following values:
Table 5 Kinetic constants obtained by fitting eqn (9) to the experimental
results for the acid hydrolysis of MNTS in the presence of a-CD, 0.168 M
HCl and variable concentrations of acetonitrile at 25.0 ◦
C
[CH3CN]/M
Kaap-pCDH/M−1
102kw/M−1 s−1
103kHCD/M−1 s−1
Scheme 6
0
47
30
30
22
17
11
8
4
1
2
3
1
2
2
3.17 0.03
3.10 0.02
3.21 0.02
3.11 0.04
3.03 0.01
2.97 0.03
2.98 0.04
8.9 0.1
5.5 0.4
5.9 0.4
3.19 × 10−2
6.38 × 10−2
0.128
The acid denitrosation of MNTS can occur in the aqueous
medium (kw) or through the MNTS–CD complex (kHCD). The bind-
6
1
a-CDH
a-CDH
MNTS
0.195
0.319
0.511
4.5 0.4
ing constants K
and K
correspond to the formation of
CH CN
3
2
3
2
3
the respective inclusion complexes between MNTS or CH3CN and
the neutral form of a-cyclodextrin. The corresponding observed
Fig. 8 (a) Influence of the concentration of a-CD on kobs for the acid hydrolysis of MNTS in the presence of 0.168 M HCl and variable concentrations
◦
of acetonitrile at 25.0 C. (᭺) [CH3CN] = 3.19 × 10−2 M; (᭹) [CH3CN] = 6.38 × 10−2 M; (ꢀ) [CH3CN] = 0.128 M; (ꢁ) [CH3CN] = 0.195 M; (ꢁ)
[CH3CN] = 0.319 M; and (ꢂ) [CH3CN] = 0.511 M. (b) Analysis of the data using eqn (9).
1046 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is
The Royal Society of Chemistry 2006
©

View Online
a-CDH
the figure to eqn (10) provided values of K
= (51 2) M−1 and
MNTS
kHCD/kw = (0.23 0.02) M−1, both of which are quite consistent
with values reported previously for this system.
The spectrophotometric results reported by Connors27 are
consistent with the formation of inclusion complexes between
a-CD
a-CD and methanol, DMSO, ethylene glycol, dioxane (K
=
1 : 1
30.6 M−1), 2-propanol, acetonitrile (K
= 40.3 M−1) and
a-CD
1 : 1
acetone, all with a 1 : 1 stoichiometry. The binding of organic
1
molecules to a-cyclodextrin has also been studied by H-NMR
spectroscopy.28 The results revealed the formation of a 1 : 1 a-
CD–acetonitrile complex and the binding constant was found to
a-CD
be K
= 5.6 M−1. A potentiometric study of the formation
1 : 1
of weak inclusion complexes between a-cyclodextrin and small
organic molecules—including ethanol, 2-propanol, 2-methyl-2-
propanol, cyclohexanol, dioxane, DMSO and phenol—in an
Fig. 9 (᭺) Variation of Kapp for the acid hydrolysis of MNTS in the
presence of a-CD and acetonitrile as a function of the concentration of
organic cosolvent. (᭹) Fitting of the results to eqn (3).
◦
a-CD
1 : 1
aqueous medium at 25.0 C provided a K
value of 5.6 M−1
in acetonitrile and 4.5 M−1 in dioxane.29 These results are quite
a-CDH
a-CDH
= (7 1) M−1], which provides
K
= (45
3) M−1 and K
= (9.0
0.6) M−1, both
value obtained in the
a-CDH
consistent with our value [K
further evidence of the accuracy of the proposed model.
MNTS
CH CN
3
CH CN
3
a-CDH
MNTS
of which are consistent with the K
absence of acetonitrile [(47 4) M−1] (see Table 5).
The kinetic results reveal the formation of a 1 : 1 complex
between acetonitrile and a-CD, which is consistent with previous
results obtained from X-ray diffraction measurements.26 Further-
more, the results for b-CD suggest the formation of a 2 : 1 complex.
This disparate behavior is a result of the increased cavity size of b-
CD relative to a-CD, which allows the former to accommodate two
acetonitrile molecules rather than one. The larger size of dioxane
relative to acetonitrile results in the formation of a 1 : 1 complex
with b-CD.
A more rigorous quantitative interpretation of the experimental
results can be provided by considering a reaction mechanism such
as that shown in Scheme 6. On the basis of this mechanism
the observed rate constant can be expressed in terms of the
concentration of free a-CD as follows:
Conclusions
The acid and basic hydrolysis reactions of MNTS in the presence
of b-cyclodextrin have been studied and the formation of an
unreactive b-CD–MNTS complex was observed in both cases.
The results with b-CD were different as a consequence of the
smaller cavity size of this cyclodextrin, which allows only partial
penetration of MNTS. The basic hydrolysis process involved
the formation of a reactive 1 : 1 a-CD–MNTS complex at low
cyclodextrin concentrations but a 2 : 1 complex was formed as the
a-CD concentration was raised.
Size differences in the host and guest molecules resulted in
different stoichiometries for the complexes. For example, dioxane
and DMSO were found to form 1 : 1 inclusion complexes with b-
and a-CD, respectively. On the other hand, acetonitrile formed 1 :
2 inclusion complexes with b-CD. The competitive formation of
such inclusion complexes accounts for the experimental behavior
observed in the acid and basic hydrolysis of MNTS in the presence
of a- or b-CD and variable amounts of organic cosolvent.
H
k
[H+] +
KMa-NCDTSH[H+][a-CD]free
a-CDH[a-CD]free
CD
kobs
kw
k
w
=
(10)
1 + KMNTS
where [a-CD]free can be obtained by using the simulation procedure
a-CDH
previously employed with dioxane [eqn (4)]. The K
obtained in this way was (7
value
CH CN
3
1) M−1 and provided the free a-
CD concentrations shown in Fig. 10. Finally, fitting the data in
Acknowledgements
Funding from Xunta de Galicia (PGIDT03-PXIC2095PN and
PGIDIT04TMT209003PR) and Spain’s Ministry of Science and
Technology (projects BQU2002-01184 and MAT2004-02991) is
gratefully acknowledged.
References and notes
1 Special issue on Cyclodextrins, ed. V. T. D’Souza and K. B. Lipkowit,
Chem. Rev., 1998, 98, 1741.
2 W. Saenger, Angew. Chem., Int. Ed. Engl., 1980, 21, 300.
3 (a) M. L. Bender and M. Komiyama, Cyclodextrin Chemistry, Springer,
New York, 1978; (b) J. Szejtli, Cyclodextrins and their Inclusion
Complexes, Akademiai Kidao, Budapest, 1982.
4 (a) R. Fornasier, M. Parmagnani and U. Tonellato, J. Inclusion Phenom.
Mol. Recognit. Chem., 1991, 11, 225; (b) A. F. Danil de Namor, R.
Traboulssi and D. F. V. Levis, J. Chem. Soc., Chem. Commun., 1990,
751; (c) N. Kobayashi and T. Osa, Carbohydr. Res., 1989, 192, 147;
(d) O. S. Tee, C. Mazza, R. Lozano-Hemmer and J. B. Giorgi, J. Org.
Chem., 1994, 59, 7602.
Fig. 10 (᭺) Influence of the concentration of free a-CD on kobs/kw for
the acid hydrolysis of MNTS in the presence of variable concentrations of
acetonitrile [data from Fig. 8 as fitted to eqn (10)]. (᭹) Results obtained in
the absence of acetonitrile.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1038–1048 | 1047
©

View Online
5 (a) A. Ueno and T. Osa, J. Inclusion Phenom., 1984, 2, 555; (b) A.
Nakajima, Bull. Chem. Soc. Jpn., 1984, 57, 1143; (c) A. E. Aboutaleb,
A. A. Abdel Rahmen and E. M. Samy, J. Pharm. Belg., 1988, 43, 437;
(d) S. Hamai, Bull. Chem. Soc. Jpn., 1992, 65, 2323; (e) Y. Liao and C.
Bohne, J. Phys. Chem., 1996, 100, 734.
6 A. S. Wang and Y. Matsui, Bull. Chem. Soc. Jpn., 1994, 67, 2917.
7 R. L. Van Etten, G. A. Clowes, J. F. Sebastian and M. L. Bender, J. Am.
Chem. Soc., 1967, 89, 3253.
15 D. Siegel and R. Breslow, J. Am. Chem. Soc., 1975, 97, 6869.
16 G. Nelson, G. Patonay and J. M. Warner, J. Inclusion Phenom., 1988,
6, 277.
17 The total concentration of MNTS was very low ([MNTS] < 5 × 10−5M),
so [MNTS] ꢀ [CD]T.
18 A. Castro, J. R. Leis and M. E. Pen˜a, J. Chem. Soc., Perkin Trans. 2,
1989, 1861.
19 (a) L. Viola and R. H. de Rossi, Can. J. Chem., 1999, 77, 860; (b) E.
8 (a) B. Siegel and R. Breslow, J. Am. Chem. Soc., 1975, 97, 6869; (b) A.
Harada and S. Takahashi, J. Inclusion Phenom., 1984, 2, 791; (c) A.
Harada and S. Takahashi, Chem. Lett., 1984, 2089.
Iglesias, J. Am. Chem. Soc., 1998, 120, 13057.
20 J. J. Taraszewska, Inclusion Phenom. Mol. Recognit. Chem., 1991, 10,
69.
9 M. Komiyama, H. Yamamoto and H. Hirai, Chem. Lett., 1984, 1081.
10 K. A. Connors, Chem. Rev., 1997, 97, 1325.
21 J. H. Park and T. H. Nah, J. Chem. Soc., Perkin Trans. 2, 1994, 1359.
22 P. K. Zarzycki and H. Lamparczyk, J. Pharm. Biomed. Anal., 1998, 18,
165.
23 L. Garc´ıa-R´ıo, J. R. Leis and J. A. Moreira, unpublished results.
24 (a) W. P. Jencks, M. T. Haber, D. Herschlag and K. L. Nazaretian, J. Am.
Chem. Soc., 1986, 108, 479; (b) J. P. Richard, J. Chem. Soc., Chem.
Commun., 1987, 1768; (c) R. A. McClelland, V. M. Kanagasabapathy,
N. S. Banait and S. Steenken, J. Am. Chem. Soc., 1992, 114, 1816;
(d) C. F. Bernasconi, K. W. Kittredge and F. X. Flores, J. Am. Chem.
Soc., 1999, 121, 6630.
11 (a) R. I. Gelb, L. M. Schwartz, M. Radeos, R. B. Edmonds and
D. A. Laufer, J. Am. Chem. Soc., 1982, 104, 6283; (b) J. Taraszewska,
J. Inclusion Phenom. Mol. Recognit. Chem., 1991, 10, 69; (c) J. Huang,
G. C. Catena and F. V. Bright, Appl. Spectrosc., 1992, 46, 606; (d) J. M.
Schuette, T. T. Ndou, A. Mun˜oz de la Pen˜a, S. Mukunden and I. M.
Warner, J. Am. Chem. Soc., 1993, 115, 292.
12 W. R. Bergmark, A. Davis, C. York, A. Macintosh and G. Jones, J. Phys.
Chem., 1990, 94, 5020.
13 (a) D. L. H. Williams, J. Chem. Soc., Perkin Trans. 2, 1976, 691; (b) L.
Garc´ıa-R´ıo, J. R. Leis, J. A. Moreira and F. Norberto, J. Chem. Soc.,
Perkin Trans. 2, 1998, 11, 756; (c) L. Garc´ıa-R´ıo, J. R. Leis, J. A. Moreira
and F. Norberto, J. Phys. Org. Chem., 1998, 11, 756.
14 (a) M. V. Rekharsky, F. P. Schwarz, Y. B. Tewari and R. N. Goldberg,
J. Phys. Chem., 1994, 98, 10282; (b) O. S. Tee and M. Bozzi, J. Am.
Chem. Soc., 1990, 112, 7815; (c) O. S. Tee, M. Bozzi, J. J. Hoeven and
T. A. Gadosy, J. Am. Chem. Soc., 1993, 115, 8990.
25 M. V. Rekharsky and Y. Inoue, Chem. Rev., 1998, 98, 1875.
26 T. Aree, J. Jacob, W. Saenger and H. Hoier, Carbohydr. Res., 1998, 307,
191.
27 K. A. Connors, M. J. Mulski and A. Paulson, J. Org. Chem., 1992, 57,
1794.
28 Y. Matsui and S. Tokunaga, Bull. Chem. Soc. Jpn., 1996, 69, 2477.
29 R. I. Gelb, L. M. Schwartz, M. Radeos, R. Edmonds and D. A. Laufer,
J. Am. Chem. Soc., 1982, 104, 6283.
1048 | Org. Biomol. Chem., 2006, 4, 1038–1048
This journal is
The Royal Society of Chemistry 2006
©