Molecular recognition study on supramolecular systems. Part 19. Circular dichroism studies of inclusion complexation of aliphatic alcohols by organoselenium modified β-cyclodextrins

Article abstract of DOI:10.1039/a808039i

Complex stability constants for the stoichiometric 1 : 1 inclusion complexation of various aliphatic alcohols with mono[6-(benzylseleno)-6-deoxy]-β-cyclodextrin (1). mono[6-(phenylseleno)-6-deoxy]-β-cyclodextrin (2), and mono-[6-(o-, m-, or p-tolylseleno)-6-deoxy]-β-cyclodextrin (3-5) have been obtained by spectrophotometric titrations at 25°C in phosphate buffer solution (pH = 7.2). The Cotton effects observed indicate that the aromatic moiety penetrates shallowly into the hydrophobia cavity of cyclodextrin. Therefore, the aromatic moiety can be taken as an induced circular dichroism (ICD) probe to investigate the inclusion phenomena. The results obtained demonstrate that the modified β-cyclodextrin (1) is highly sensitive to the size/shape and conformational rigidity of guest molecules, giving fairly good molecular selectivity up to 114 for adamantan-1-ol/cyclopentanol and relatively high EIZ selectivity up to 2.7 for geraniol/nerol. Interestingly, all of the modified β-cyclodextrins employed displayed relatively good enantioselectivity for (+)-enantiomers of borneol and menthol. The molecular recognition ability and enantioselectivity for aliphatic alcohols of the modified β-cyclodextrins (1-5) are discussed from the viewpoints of the size/shape-fit relationship and the multipoint recognition mechanism.

Full text of DOI:10.1039/a808039i

View Article Online / Journal Homepage / Table of Contents for this issue
Molecular recognition study on supramolecular systems. Part 19.†
Circular dichroism studies of inclusion complexation of aliphatic
alcohols by organoselenium modified â-cyclodextrins
Yu Liu,*ab Bin Li,a Bao-Hang Han,ac Takehiko Wadac and Yoshihisa Inoue*bc
a Department of Chemistry, Nankai University, Tianjin 300071, China
b Inoue Photochirogenesis Project, ERATO, JST, 4-6-3 Kamishinden, Toyonaka 565-0085,
Japan
c Department of Molecular Chemistry, Osaka University, Yamadaoka, Suita 565-0871, Japan
Received (in Cambridge) 16th October 1998, Accepted 22nd December 1998
Complex stability constants for the stoichiometric 1:1 inclusion complexation of various aliphatic alcohols with
mono[6-(benzylseleno)-6-deoxy]-β-cyclodextrin (1), mono[6-(phenylseleno)-6-deoxy]-β-cyclodextrin (2), and mono-
[6-(o-, m-, or p-tolylseleno)-6-deoxy]-β-cyclodextrin (3–5) have been obtained by spectrophotometric titrations at
25 ЊC in phosphate buffer solution (pH = 7.2). The Cotton effects observed indicate that the aromatic moiety
penetrates shallowly into the hydrophobic cavity of cyclodextrin. Therefore, the aromatic moiety can be taken as an
induced circular dichroism (ICD) probe to investigate the inclusion phenomena. The results obtained demonstrate
that the modified β-cyclodextrin (1) is highly sensitive to the size/shape and conformational rigidity of guest
molecules, giving fairly good molecular selectivity up to 114 for adamantan-1-ol/cyclopentanol and relatively high
E/Z selectivity up to 2.7 for geraniol/nerol. Interestingly, all of the modified β-cyclodextrins employed displayed
relatively good enantioselectivity for (ϩ)-enantiomers of borneol and menthol. The molecular recognition ability and
enantioselectivity for aliphatic alcohols of the modified β-cyclodextrins (1–5) are discussed from the viewpoints of
the size/shape–fit relationship and the multipoint recognition mechanism.
Introduction
Molecular recognition by modified cyclodextrins is currently a
significant topic in supramolecular chemistry. Modification of
native cyclodextrins by introducing nucleophilic or electrophilic
substituents can alter not only the original molecular binding
ability but also the relative molecular selectivity and enantio-
selectivity.1–6 Consequently, a wide variety of cyclodextrin
derivatives have been designed and synthesized in order to
investigate the recognition mechanism controlled by the simul-
taneous operation of several weak interactions and the factors
governing inclusion phenomena of guest molecules by host
cyclodextrins.7–21 Aliphatic alcohols with a variety of structures
have been employed as model guest compounds to study sys-
tematically the inclusion complexation behavior of natural and
modified cyclodextrins. Matsui et al.22 reported the inclusion
complexation thermodynamics of natural cyclodextrins with
aliphatic alcohols. More recently, Ueno et al.23 have reported
the molecular recognition of aliphatic alcohols by modified
cyclodextrins carrying a dansyl moiety as fluorescent probe,
giving interesting results. These results indicated that, in
addition to the size and shape of aliphatic alcohols, the micro-
structural change of the cyclodextrins apparently governs the
inclusion complexation phenomena to some extent. Therefore,
the elucidation of the inclusion mechanism is also helpful for
our further understanding of the multipoint recognition and the
induced-fit interaction hypothesis proposed for the selective
binding of specific substrate by biological receptors.
In the present study, we report our synthesis of the series of
arylseleno derivatives of β-cyclodextrin (1–5) shown in Chart 1,
and investigation of their inclusion complexation with selected
aliphatic alcohols in phosphate buffer solution (pH = 7.2) at
25 ЊC using differential circular dichroism spectrometry.24
Chart 1
One important reason for choosing organoselenium-modified
cyclodextrins as hosts is that selenium, possessing a larger
atomic radius and lower electronegativity than carbon, can lead
to a C–Se bond which is longer and more flexible than a C–C
bond. The series of aliphatic alcohols shown in Chart 2 are
employed as guest molecules in order to examine the possible
participation of several weak interactions working in the com-
plexation with the organoselenium-modified β-cyclodextrins.
Under such circumstances, we can discuss the molecular recog-
nition ability and enantioselectivity upon complexation of
aliphatic alcohols by β-cyclodextrin derivatives 1–5 in terms of
the size–fit and the complementary geometrical relationship
between the host cyclodextrins and guest aliphatic alcohols.
The complex stability constant (log Ks) and Gibbs free energy
changes (Ϫ∆GЊ) obtained will provide further understanding
† For Part 18, see ref. 30.
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568
563

View Article Online
Chart 2
of the potential importance of such studies in discussing the
relevant stereochemical complementary geometrical relation-
ship between the biological receptor (host) and substrate (guest)
interaction.1
Fig. 1 Circular dichroism spectra of β-cyclodextrin derivatives 1 and 2
(top) and 3–5 (bottom) (6 × 10Ϫ5 mol dmϪ3) in phosphate buffer solu-
tion at 25.0 ЊC.
Experimental
Spectroscopy
The precipitate formed was filtered to give a white powder. The
crude product was purified by three recrystallizations from
water and dried in vacuo to give a pure sample (yield 50%). 1H
NMR (200 MHz, [2H6]DMSO, 25 ЊC, TMS): δ 3.1–3.9 (m,
40H), 4.1–4.6 (m, 8H), 4.8–5.2 (9H), 5.3–5.8 (m, 14H), 7.3
(m, Ar 5H). IR (KBr) ν/cmϪ1 3369.0, 2912.5, 1730.9, 1701.5,
1638.6, 1615.4, 1574.6, 1536.7, 1514.6, 1398.6, 1365.0, 1336.5,
1302.9, 1273.9, 1235.1, 1149.4, 1127.1, 1073.8, 1022.3, 938.4,
885.6, 789.7, 769.3, 692.7, 658.7. UV/vis (water) λmax/nm (ε/dm3
molϪ1 cmϪ1) 260.2 (613). C49H76O34Seؒ6H2O (1396.2): calcd. C,
42.15; H, 6.35. Found: C, 42.12; H, 6.10%.
Circular dichroism (CD) spectra were recorded on a JASCO
J-720 spectropolarimeter.
Materials
Guest aliphatic alcohols employed were commercially available
and were used as received. Adamantan-1-ol, cyclohexanol and
cyclooctanol were purchased from Nacalai Tesque, Inc.
Cyclopentanol, (Ϫ)-borneol, (ϩ)-menthol, (Ϫ)-menthol, and
nerol were purchased from Tokyo Kasei. (ϩ)-Borneol, cyclo-
heptanol, and geraniol were purchased from Aldrich, Merck
and Wako, respectively. A series of organoselenium-modified
β-cyclodextrins, bearing benzylseleno (1), phenylseleno (2), and
o-, m- and p-tolylseleno (3–5) groups were synthesized in 40–
55% yields, respectively, by the reaction of mono[6-O-(p-tolyl-
sulfonyl)]-β-cyclodextrin with the corresponding aromatic
selenide anions in DMF, according to the procedures reported
recently.25 A representative synthetic procedure and charac-
terization of modified cyclodextrin (1) is as follows.
Mono[6-O-(p-tolylsulfonyl)]-β-cyclodextrin (6-OTs-β-CD)
was prepared by a reaction of β-cyclodextrin with toluene-p-
sulfonyl chloride in dry pyridine.26 Compound 1 was synthesized
by the reaction of mono[6-O-(p-tolylsulfonyl)]-β-cyclodextrin
(6-OTs-β-CD) with dibenzyl diselenide27 according to the
following procedure. Sodium borohydride (0.037 g, 1 mmol)
was added to the yellow solution of dibenzyl diselenide (0.17 g,
0.5 mmol) in dry ethanol (50 ml) with stirring under nitrogen at
room temperature. Once the solution became colorless, a solu-
tion of mono[6-O-(p-tolylsulfonyl)]-β-cyclodextrin (1.29 g, 1
mmol) in dry DMF (75 ml) was added dropwise into the solu-
tion, which was then heated to 60 ЊC for 2 h with stirring. The
resultant solution was evaporated under reduced pressure to
give a light yellow powder, which was then dissolved in a min-
imum amount of hot water, and poured into acetone (100 ml).
Spectral measurements
The binding constants for the inclusion complexation of
modified β-cyclodextrins (1–5) with some selected aliphatic
alcohols were determined using differential circular dichroism
(CD) spectrometry. The sample solutions containing modified
β-cyclodextrins 1–5 (0.06 mmol dmϪ3) and varying concentra-
tions of guests (0.6–9.6 mmol dmϪ3) were kept at 25.0 0.1 ЊC
by circulating thermostatted water through the jacket. The dif-
ferential CD spectrum was obtained by subtracting the original
CD spectrum in the absence of guest from that in the presence
of a guest on computer memories. Simultaneous examination
of each sample solution by UV spectrometry did not show any
significant change upon addition of the guests.
Results and discussion
CD Spectra
As can be seen from Fig. 1, the CD spectrum of modified β-
cyclodextrin 2 in aqueous solution showed a strong negative
1
Cotton effect peak, corresponding to the La band, at 232 nm
(∆ε = Ϫ3.41) and a weak positive Cotton effect for the 1Lb band
at 279 nm (∆ε = 0.99). The other modified cyclodextrins also
1
showed such a strong negative Cotton effect peak for the La
564
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568

View Article Online
Fig. 4 Curve fitting analyses, according to eqn. (2), for complexations
of cyclooctanol, (ϩ)-borneol, and (Ϫ)-borneol with 1 at 25.0 ЊC.
can be used to determine the complex stability constants.
Assuming 1:1 stoichiometry, the inclusion complexation of
aliphatic alcohols (G) with modified β-cyclodextrins (H) is
expressed by eqn. (1).
1
Fig. 2 Kajtar’s sector rule applied to transition moments of La and
1Lb bands of the phenylseleno moiety in modified β-cyclodextrin 2.
Ks
(1)
H ϩ G
GؒH
The stability constant (Ks) of the inclusion complex formed
can be determined using a non-linear least squares method
according to the curve fitting eqn. (2),29 where [G]0 and [H]0
∆∆ε = {α([H]0 ϩ [G]0 ϩ 1/Ks)
√α2([H]0 ϩ [G]0 ϩ 1/Ks)2 Ϫ 4α2[H]0[G]0}/2 (2)
refer to the total concentration of aliphatic alcohols and
β-cyclodextrin derivatives, respectively; α is the proportionality
coefficient, which may be taken as a sensitivity factor for the
CD change; ∆∆ε denotes the change in the CD spectrum of
β-cyclodextrin derivative upon stepwise addition of the guest
alcohol. For each host compound examined, the plot of ∆∆ε as
a function of [G]0 gave an excellent fit, verifying the validity of
the 1:1 complex stoichiometry assumed above. As shown in
Fig. 4, where the ∆∆ε values are plotted against the [G]0 values
to give an excellent fit, no serious deviations are found in the
curve fitting. When repeated measurements were made, the Ks
value was reproducible within an error of 5%, which corre-
sponds to an estimated error of 0.15 kJ molϪ1 in the free energy
of complexation (∆GЊ). The complex stability constants (Ks)
and the sensitivity factor α obtained by the curve fitting are
listed in Table 1, along with the free energy change of complex
formation (Ϫ∆GЊ).
Fig. 3 The circular dichroism spectrum of β-cyclodextrin derivative
1 in the presence of cyclooctanol at various concentrations at 25.0 ЊC.
The concentration of 1 is 6 × 10Ϫ5 mol dmϪ3. The concentrations
of cyclooctanol increase in the range of 0–9.6 × 10Ϫ3 mol dmϪ3 from
a to g.
band and a weak positive Cotton effect for the 1Lb band.
According to the sector rule proposed by Kajtar et al.,28 these
negative and positive Cotton effects observed respectively for
1
1
the La and Lb bands indicate that the aromatic moiety pene-
trates shallowly into the hydrophobic cavity of cyclodextrin. As
illustrated schematically in Fig. 2, the transition moment of the
1
1La band lies in the negative region, while that of the Lb band
is in the positive region, coinciding with the Cotton effect
observed. It is thus inferred that the arylseleno moiety is only
shallowly included in the cyclodextrin cavity. Hence, we can use
the aromatic moiety as a CD probe to investigate the inclusion
behavior.
Molecular recognition ability and enantioselectivity
Extensive studies on molecular recognition by cyclodextrins2
have shown that an important characteristic of the complex-
ation is the simultaneous operation of several weak forces
working between the guest and host. In the present case, the
relative size and stereochemical complementary relationships
between host and guest, hydrogen-bonding and hydrophobic
interactions, and the induced dipole of the functional sidearm
perching on the edge of cyclodextrin cavity are considered to be
important. The results obtained indicate that the self-inclusion
of the β-cyclodextrin’s sidearm plays an important role in
determining how the guest molecule fits into the host cavity,
according to the sidearm’s size, shape, dipole, charge, and func-
tional group.
CD spectral titrations
As can be seen from Fig. 3, in the titration experiments using
differential CD spectrometry, gradual addition of a known
concentration of cyclooctanol to a dilute host solution (0.06
mmol dmϪ3) in phosphate buffer solution caused significant
1
decreases in intensity in the La band, while the change in the
1Lb band was minimal at this host concentration. This result
indicates that the aromatic moiety, initially perching on the edge
of the cyclodextrin cavity, suffers substantial conformational
changes upon guest inclusion, probably moving out of the
chiral hydrophobic cavity, This substantial CD spectral change
As can be seen from Table 1, all of the modified β-cyclo-
dextrins except 2 show relatively high molecular binding ability
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568
565

View Article Online
Table 1 Complex stability constants (log Ks), Gibbs free energy
changes (Ϫ∆GЊ) and sensitivity factors α for the inclusion complexation
of modified β-cyclodextrins 1–5 with aliphatic alcohols in phosphate
buffer solution (pH = 7.20) at 25 ЊC
Ϫ∆GЊ/
Host
Guest
Ks
log Ks
kJ molϪ1
α/102
1
Cyclopentanol
Cyclohexanol
Cycloheptanol
Cyclooctanol
Adamantan-1-ol
Geraniol
68
138
470
996
7760
857
317
887
835
1.83
2.14
2.67
3.00
3.89
2.93
2.50
2.95
2.92
3.76
3.64
2.59
3.22
3.36
3.24
4.42
4.32
3.81
3.33
3.30
4.59
3.65
3.39
4.16
10.5
12.2
15.3
17.1
22.2
16.7
14.3
16.8
16.7
21.5
20.7
14.8
18.4
19.2
18.5
25.2
24.6
21.8
19.0
18.8
26.2
20.8
19.4
23.7
730
815
869
962
1020
538
1060
800
929
924
1130
Nerol
(ϩ)-Menthol
(Ϫ)-Menthol
(ϩ)-Borneol
(Ϫ)-Borneol
Cyclooctanol
Cyclooctanol
(ϩ)-Menthol
(Ϫ)-Menthol
(ϩ)-Borneol
(Ϫ)-Borneol
Cyclooctanol
(ϩ)-Menthol
(Ϫ)-Menthol
(ϩ)-Borneol
Cyclooctanol
(ϩ)-Menthol
(ϩ)-Borneol
5740
4350
386
2
3
35.1
Fig. 5 Gibbs free energy changes (Ϫ∆GЊ) of complexation of cyclo-
alkanols (C5–C8), adamantan-1-ol (1-Ad), (ϩ)- and (Ϫ)-borneol
(Bor), (ϩ)- and (Ϫ)-menthol (Men), geraniol (Ger), and Nerol (Ner)
with mono[6-(benzylseleno)-6-deoxy]-β-cyclodextrin as a function of
number of carbon atoms (NC) in the guest molecule.
1670
2270
1720
26300
20700
6520
2140
2000
39200
4440
2470
14300
2010
1630
1820
1900
1910
1030
778
658
1030
452
4
5
111
668
to borneol and adamantan-1-ol, which may be attributed to the
strict size–fit relationship between the host and the spherical
guests. The aromatic substituent on the sidearm attached to the
edge of cyclodextrin also plays a crucial role in guest inclusion.
Thus, the Ks value for cyclooctanol increases from 386 for 2
(phenyl) to 996 for 1 (benzyl) and then to 1670–6520 (tolyl)
for 3–5 with increasing hydrophobicity of the substituent
introduced. This clearly indicates that the aromatic substituent,
initially perching in the cavity in the absence of the guest, is
driven out of the cavity upon guest inclusion, expanding the
original hydrophobic cavity to the primary side.
As can be readily recognized from the data for 1 (Table 1),
the binding constant is highly sensitive to the size/shape and
rigidity of the guest molecules, giving moderate molecular selec-
tivity up to 14.6 for cyclooctanol/cyclopentanol and fairly good
E/Z selectivity up to 2.7 for geraniol/nerol. For the hosts 3–5,
the combined effect of size/shape–fit and substituent led to the
enhanced molecular recognition ability and isomer selectivity.
However, most of the host compounds display only relatively
low enantioselectivities of 1.1–1.3. It seems reasonable that
modified β-cyclodextrins show high recognition ability for
guests possessing distinctly different size/shape, but only low
enantioselectivity for enantiomers possessing small structural
differences.
Fig. 6 Sensitivity factor (α) and Gibbs free energy changes (Ϫ∆GЊ)
upon complexation of cycloalkanols (C5–C8), adamantan-1-ol (1-Ad),
geraniol (Ger), Nerol (Ner), (ϩ)- and (Ϫ)-menthol (Men), and (ϩ)- and
(Ϫ)-borneol (Bor) with mono[6-(benzylseleno)-6-deoxy]-β-cyclodextrin
1; compare the different changing profiles of α and Ϫ∆GЊ for the
terpenoid alcohols.
in Fig. 5. This result may be attributed to less-efficient van der
Waals interactions of these acyclic or branched C10 alcohols as
well as the larger entropic losses of acyclic alcohols upon
complexation.
It is also interesting to compare the changing profiles of
Ϫ∆GЊ value and of sensitivity factor α, both of which are
shown in Fig. 6. As far as the cycloalkanols and adamantanol
are concerned, both values behave quite similarly, indicating
that the larger guests, which are bound more strongly, induce
more extensive conformational changes. This result seems to
imply that the α value can be taken as a quantitative measure of
the conformational change induced by guest inclusion in the
host cavity, at least for the cycloalkanol series. However, this
idea cannot immediately be extended to the complexation
behavior of the other guest series even in a qualitative sense,
since the highest Ks value does not always accompany the most
drastic CD change or the highest α value with the acyclic and
more complicated alcohols employed. This apparent discrep-
ancy may be rationalized by taking into account that the CD
spectral change only represents fairly local conformational
changes around the chromophore, while the binding constant
reflects more global changes in the cooperative weak inter-
actions between the host and guest. Hence, these results
are helpful to our further understanding of the multipoint
Mono[6-(benzylseleno)-6-deoxy]-â-cyclodextrin (1)
In order to evaluate the inclusion complexation behavior of 1
with the aliphatic alcohols from a more quantitative point of
view, the free energy change (Ϫ∆GЊ) was plotted against the
formal number of carbon atoms (NC) in the guest molecule
in Fig. 5. As has been observed frequently with native and
modified cyclodextrins,2 the plot gives a good straight line at
least with the cycloalkanols to give a unit increment in ∆GЊ
per methylene of 2.2 kJ molϪ1, which is somewhat smaller
than the typical unit increment of 2.8 0.8 kJ molϪ1 observed
in the complexation of a wide variety of cycloalkanols with
β-cyclodextrin.2 It is noted that the plot for adamantan-1-ol
falls exactly on the regression line when it is treated as a C10
cycloalkanol. In contrast, the other terpenoid alcohols employed
(formal NC = 10), gave much lower complex stabilities, as shown
566
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568

View Article Online
recognition mechanism. In general, the value of sensitivity
factor α only represents the microstructural change on inclusion
complexation, which mainly depends on the depth of guest
penetration into the cavity, which is, however, not the only
factor that influences the inclusion complexation. It has been
extensively verified that the stability of an inclusion complex
with modified cyclodextrin is governed by several cooperative
weak forces working between host and guest, including hydro-
phobic, hydrogen-bonding, dipole–dipole, electrostatic and
van der Waals interactions.2 Therefore, it is not particularly
curious that the sensitivity factor α shows only a poor positive
correlation (correlation coefficient r 0.46, for the inclusion
complexation of host 1 with the aliphatic alcohols examined)
with the complex stability.
It is more interesting that host 1 shows the strongest binding
ability for adamantan-1-ol and gives the highest molecular
selectivity up to 114 for adamantan-1-ol/cyclopentanol. One
reasonable explanation is that adamantan-1-ol, possessing a
large rigid hydrophobic spherical structure, can induce the
strongest hydrophobic interaction and establish the best size–fit
relationship between the host and guest. Furthermore, owing
to its most rigid and hydrophobic structure, adamantan-1-ol
forms the most stable complex with modified β-cyclodextrin 1
of the seven C10 aliphatic alcohols.
As can be seen from Table 1, modified β-cyclodextrin 1 can
recognize not only the size of the aliphatic alcohols, but also the
shape and chirality of the isomers. For borneol, menthol, nerol
and geraniol, possessing obviously different shape and rigidity,
modified β-cyclodextrin 1 displays fairly good isomer separ-
ation: (ϩ)-borneol > (Ϫ)-borneol > geraniol > (ϩ)-menthol >
(Ϫ)-menthol > nerol, giving highest isomer selectivity up to
18.1 for (ϩ)-borneol/nerol. This order of complex stability
is mostly determined by the rigidity and size/shape of the
guests. Therefore, possessing the most rigid molecular structure,
borneol forms the most stable complex with host 1, while the
most flexible, least bulky nerol with an (E)-double bond gives
the least stable complex. It is somewhat unexpected that
geraniol and menthol form complexes of comparable stability,
which might be attributable to the more bulky structure
imposed by the (E)-double bond. Host 1 shows relatively low
enantioselectivity for borneol, the enantioselectivity calculated
from the Ks values is 1.3 for (ϩ)-borneol/(Ϫ)borneol. However,
owing to their distinctly different shape, the geometrical iso-
mers, nerol and geraniol, show substantially different Ks values;
the E/Z selectivity calculated from the Ks values amount to 2.7
for geraniol/nerol.
hydrophobic expanded cavity. These hosts display enhanced
molecular recognition ability and enantioselectivity for borneol
and menthol over 1 or 2. Host 3 shows relatively high isomer
selectivity calculated from the Ks values up to 15.3 for
(ϩ)-borneol/(Ϫ)-menthol and relatively good enantioselectivi-
ties of 1.3 for (ϩ)-borneol/(Ϫ)-borneol and 1.3 for (ϩ)-menthol/
(Ϫ)-menthol. Similarly, host 4 shows fairly good isomer select-
ivity up to 19.6 for (ϩ)-borneol/(Ϫ)-menthol. However, host 5
displays much lower isomer selectivity than 3 and 4, giving
the isomer selectivity of 5.8 for (ϩ)-borneol/(ϩ)-menthol.
Although the hosts 3–5, possessing isomeric tolyl substituents,
generally exhibit similar binding constants for most guest alco-
hols, the o-tolyl host 3 gives a much smaller Ks value only for
cyclooctanol and higher enantioselectivities of 1.3 for borneol
and menthol. These somewhat puzzling results would be
accounted for in terms of the original penetration of the
o-methyl group in the cavity, which interferes with the inclusion
of larger-sized cyclooctanol but discriminates more precisely
the enantiomeric isomers included in the cavity.
Conclusions
The present study indicates that a series of modified β-cyclo-
dextrins possessing a single arylseleno moiety as a CD probe
can recognize minimal differences between aliphatic alcohols
based on their size, shape, rigidity and chirality. The chromo-
phoric probe perching on the edge of the β-cyclodextrin cavity
can produce conformational change induced by guest inclusion,
which is useful in determining complex stability constants.
Especially, the benzylseleno moiety of cyclodextrin derivative
1 is most sensitive to the microstructural difference of guest
molecules among the hosts 1–5. Although all of the modified
β-cyclodextrins show low to moderate enantioselectivities for
( )-borneol and ( )-menthol, they display relatively high
molecular recognition ability for cyclic alcohols and fairly good
isomer separation and E/Z selectivity for C10 alcohols. More-
over, the host compounds 3–5 bearing a tolylseleno moiety can
enhance both molecular binding ability and selectivity. Experi-
mentally, adamantan-1-ol and (ϩ)-borneol, possessing the
most rigid and hydrophobic structures, have the best-fitted size
and shape among the aliphatic alcohols examined. These results
demonstrate that the size/shape–fit, induced-fit, and substituent
effect as well as the multipoint recognition mechanism play
crucial roles in the inclusion complexation.
Acknowledgements
This work was supported by the National Outstanding Youth
Fund (Grant No. 29625203) and Natural Science Foundation
(Grant No. 29676021) of China, Tianjin Natural Science Fund
(Grant No. 973602211) and Transcentury Qualified Personal
Fund of Tianjin Education Committee (Sun-light Plan), and
of State Education Committee of China, which are gratefully
acknowledged.
Mono[6-(phenylseleno)-6-deoxy]-â-cyclodextrin (2), mono-
[6-(o-tolylseleno)-6-deoxy]-â-cyclodextrin (3), mono[6-(m-tolyl-
seleno)-6-deoxy]-â-cyclodextrin (4) and mono[6-(p-tolylseleno)-
6-deoxy]-â-cyclodextrin (5)
As can be seen from Table 1, the five modified β-cyclodextrins
display drastically different Ks values for the inclusion complex-
ation with cyclooctanol: i.e. 4 > 5 > 3 > 1 > 2. As compared
with the modified β-cyclodextrin 2, the isomers of 1, 3, 4 and 5,
possessing a methylene or methyl substitution in the arylseleno
sidearm attached to the edge of the β-cyclodextrin, must pro-
duce substantially different conformational change induced by
the same guest, which can be further proved by the drastic
change of the values of the sensitivity factor α. Somewhat
unexpectedly, m-isomer 4 displays much higher, nearly four-fold
calculated from Ks values, binding stability for cyclooctanol than
o-isomer 3. One possible explanation for the enhanced binding
ability of host 4 with cyclooctanol is that the self-inclusion of
the m-tolyl moiety attached to the primary edge of 4 caused
more strict complementary geometrical relationship between
the cavity of modified β-cyclodextrin 4 and cyclooctanol.
References
1 Part 13: Y. Liu, B. Li, B.-H. Han, Y.-M. Li and R.-T. Chen, J. Chem.
Soc., Perkin Trans. 2, 1997, 1275.
2 M. V. Rekharsky and Y. Inoue, Chem. Rev., 1998, 98, 1875.
3 H. Ikeda, M. Nakamura, N. Ise, N. Oguma, A. Nakamura, T. Ikeda,
F. Toda and A. Ueno, J. Am. Chem. Soc., 1996, 118, 10980.
4 K. Hamasaki, H. Ikeda, A. Nakamura, A. Ueno, F. Toda, I. Suzuki
and T. Osa, J. Am. Chem. Soc., 1993, 115, 5035.
5 K. A. Martin, M. A. Mortellaro, R. W. Sweger, L. E. Fikes, D. T.
Winn, S. Clary, M. P. Johnson and A. W. Czarnik, J. Am. Chem.
Soc., 1995, 117, 10443.
6 R. M. Tarkka and E. Buncel, J. Am. Chem. Soc., 1995, 117, 1503.
7 E. U. Akkaya and A. W. Czarnik, J. Am. Chem. Soc., 1988, 110,
8553.
As for borneol and menthol, hosts 3–5 show much higher
molecular binding ability than 1, probably due to the more
8 L. E. Fikes, D. T. Winn, R. W. Sweger, M. P. Johnson and A. W.
Czarnik, J. Am. Chem. Soc., 1992, 114, 1493.
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568
567

View Article Online
9 A. Harada, Coord. Chem. Rev., 1996, 148, 115.
10 L. Jullien, J. Canceill, B. Valeur, E. Bardez, J.-P. Lefèvre, J.-M. Lehn,
V. Marchi-Artzner and R. Pansu, J. Am. Chem. Soc., 1996, 118,
5432.
11 L. Jullien, J. Canceill, B. Valeur, E. Bardez and J.-M. Lehn, Angew.
Chem., Int. Ed. Engl., 1994, 33, 2438.
12 M. T. Rojas, R. Koniger, J. F. Stoddart and A. E. Kaifer, J. Am.
Chem. Soc., 1995, 117, 336.
13 G. Nelles, M. Weisser, R. Back, P. Wohlfart, D. Wenz and S. Mittler-
Neher, J. Am. Chem. Soc., 1996, 118, 5039.
14 T. V. S. Rao and D. S. Lawrence, J. Am. Chem. Soc., 1990, 112, 3614.
15 J. S. Manka and D. S. Lawrence, J. Am. Chem. Soc., 1990, 112, 2440.
16 G. Li and L. B. McGown, Science, 1994, 264, 249.
17 A. Ueno, T. Kuwabara, A. Nakamura and F. Toda, Nature, 1992,
356, 136.
18 A. Ueno, I. Suzuki and T. Osa, J. Am. Chem. Soc., 1989, 111, 6391.
19 R. Breslow, Science, 1982, 218, 532.
22 Y. Matsui and K. Mochida, Bull. Chem. Soc. Jpn., 1979, 52, 2808.
23 H. Ikeda, M. Nakamum, N. Ise, F. Toda and A. Ueno, J. Org.
Chem., 1997, 62, 1414.
24 L.-H. Tong, Z.-J. Hou, Y. Inoue and A. Tai, J. Chem. Soc., Perkin
Trans. 2, 1992, 1253.
25 Y. Liu, B. Li, T. Wada and Y. Inoue, Supramol. Chem., 1999, in
press.
26 Y. Matsui, T. Yokoi and K. Mochida, Chem. Lett., 1976, 1037.
27 W. H. H. Gunther, J. Org. Chem., 1967, 32, 3929.
28 M. Kajtar, C. Harvath-Toro, E. Kuthi and J. Szejtli, Acta Chim.
Acad. Sci. Hung., 1982, 110, 327.
29 Y. Inoue, K. Yamamoto, T. Wada, S. Everitt, X.-M. Gao, Z.-J. Hou,
L.-H. Tong, S.-K. Jiang and H.-M. Wu, J. Chem. Soc., Perkin Trans.
2, 1998, 1807.
30 Part 18. Y. Liu, B. Li, T. Wada and Y. Inoue, Supramol. Chem., 1999,
in the press.
20 I. Tabushi, Tetrahedron, 1984, 40, 269.
21 A. P. Croft and R. A. Bartsch, Tetrahedron, 1983, 39, 1417.
Paper 8/08039I
568
J. Chem. Soc., Perkin Trans. 2, 1999, 563–568