Solubilization of polycyclic aromatics in water by γ-cyclodextrin derivatives

Article abstract of DOI:10.1002/asia.201100217

A series of hydrophilic per-6-thio-6-deoxy-γ-cyclodextrins (CDs) were synthesized from per-6-iodo-6-deoxy-γ-CD. These new hosts are able to solubilize polycyclic aromatic guests in aqueous solution to much higher extents than native CDs. Phase-solubility diagrams were mostly linear in accordance with both 1:1 and 1:2 CD-guest complexes in aqueous solution. The stoichiometry of the inclusion complexes was further investigated by fluorescence spectroscopy, which revealed very pronounced Stokes shifts typical for 1:2 complexes. This finding was further consolidated by quantum mechanical calculations of dimer formation of the guests and space-filling considerations by using the cross-sectional areas of the CDs and guests. The calculated dimerization energies correlated well with the binding free energies measured for the 1:2 complexes, and provided the main contribution to the driving force of complexation in the γ-CD cavity. Copyright

Full text of DOI:10.1002/asia.201100217

FULL PAPERS
DOI: 10.1002/asia.201100217
Solubilization of Polycyclic Aromatics in Water by g-Cyclodextrin
Derivatives
Hai Ming Wang and Gerhard Wenz*[a]
Abstract: A series of hydrophilic per-6-
thio-6-deoxy-g-cyclodextrins (CDs)
plexes was further investigated by fluo-
rescence spectroscopy, which revealed
very pronounced Stokes shifts typical
for 1:2 complexes. This finding was fur-
ther consolidated by quantum mechan-
ical calculations of dimer formation of
the guests and space-filling considera-
tions by using the cross-sectional areas
of the CDs and guests. The calculated
dimerization energies correlated well
with the binding free energies mea-
sured for the 1:2 complexes, and pro-
vided the main contribution to the
driving force of complexation in the g-
CD cavity.
were synthesized from per-6-iodo-6-
deoxy-g-CD. These new hosts are able
to solubilize polycyclic aromatic guests
in aqueous solution to much higher ex-
tents than native CDs. Phase-solubility
diagrams were mostly linear in accord-
ance with both 1:1 and 1:2 CD–guest
complexes in aqueous solution. The
stoichiometry of the inclusion com-
Keywords: cyclodextrins · dimeri-
zation
· fluorescence · inclusion
compounds · polycycles
Introduction
Composed of a hydrophilic shell and a relatively hydro-
phobic cavity, CDs solubilize hydrophobic organic com-
pounds through the formation of water-soluble inclusion
complexes mainly driven by hydrophobic interaction.[3c,5]
There are indeed many reports on the solubility enhance-
ments of hydrophobic or amphiphilic guests, including phar-
maceutical drugs, by CDs in aqueous solution.[3c,6] But, due
to the limited water solubilities of native CDs, the solubility
enhancements reached by native CDs are generally small.
However, modifications on the free hydroxyl groups of
native CDs with appropriate functional groups[5b,7] may not
only significantly increase the water solubility of CDs, but
also greatly increase the ability of CDs to form inclusion
complexes with guest molecules. For instance, methylated b-
CD and 2-hydroxypropyl-b-CD are superior in the solubili-
zation of drugs relative to native b-CD.[5b] CD sulfonates[8]
perform even better than the neutral derivatives, especially
for the solubilization of hydrophobic guests such as naphtha-
lene (NAP) in water.[7a] Recently, we found a new class of
hydrophilic per-6-deoxy-thioethers of b-CD, which showed
exceptionally high binding constants (K>106 mꢀ1, equivalent
to DG<ꢀ34 kJmolꢀ1) for para-disubstituted benzene deriv-
atives in water. The higher hydrophobicity of sulfur com-
pared to oxygen, which leads to an extension of the hydro-
phobic cavity, and additional Coulomb interactions, were
supposed to be responsible for the observed high binding
potentials of these hosts.[2c,9]
Cyclodextrins (CDs) are a series of cyclic oligosaccharides.
They consist of a-1,4-linked glucose subunits that form
hollow truncated cones of various diameters.[1] Those CDs,
composed of six, seven, and eight glucose units, are pro-
duced on industrial scales and called a-, b-, and g-CDs, re-
spectively. On account of their unique molecular structure,
CDs and their derivatives have the remarkable ability to ac-
commodate a wide variety of guest molecules in aqueous so-
lution by means of noncovalent interactions.[2] This property
is currently used in numerous applications in pharmaceuti-
cal, food, and chemical industries, and is also extensively ex-
ploited to improve the efficiency and selectivity of chemical
reactions.[3] Many efforts are now dedicated to the improve-
ment of solubility, stability, reactivity, and other properties
of guest molecules by complexation in CDs and CD deriva-
tives.[2b,4]
[a] H. M. Wang, Prof. Dr. G. Wenz
Organische Makromolekulare Chemie
Saarland University
Geb. C4.2, 66123 Saarbrꢀcken (Germany)
Fax : (+49)681-302-3909
E-mail: g.wenz@mx.uni-saarland.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100217.
2390
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2390 – 2399

Only a few efforts have presently been made to evaluate
the capabilities of modified g-CD for solubility enhance-
ments of guest molecules.[10] It is known that g-CD is able to
accommodate two aromatic guests simultaneously within its
relatively large cavity, thereby leading to an inclusion com-
pound with 1:2 stoichiometry.[2b] Consequently, g-CD and its
derivatives are very interesting nanovessels for the execu-
tion of various bimolecular reactions, especially cycloaddi-
tions.[11] For instance, Nakamura and Inoue reported that
the host–guest ratio in the inclusion complex of g-CD with
2-anthracene carboxylate in water was 1:2 based on the re-
sults from ROESY-NMR spectroscopy measurements.[12]
Also the ROESY-NMR spectrum of the inclusion system of
g-CD with methyl 3-methoxyl-2-naphthoate demonstrated
the existence of a complex with 1:2 stoichiometry.[11c] Ac-
cordingly, regio- and stereoselective photodimerizations be-
tween two substrates, both in solution and in the solid state,
had been achieved in the presence of g-CD.[11a,k]
In this work we report on the synthesis of highly water-
soluble thioethers of g-CD with flexible neutral or ionic side
arms, because we hoped that this derivatization might lead
to a similar enhancement of the binding potential like that
observed for b-CD thioethers.[2a,c]
Scheme 1. Chemical structures of the selected polycyclic aromatic guests:
naphthalene, NAP; 2-naphthalenecarboxylic acid, NCA; azulene, AZU;
trans-stilbene, STI; acenaphthylene, ACE; anthracene, ANT; phenan-
threne, PHE; tetracene, TET.
Eight polycyclic aromatics of similar size (see Scheme 1):
napthalene (NAP), 2-naphtha-
lenecarboxylic acid (NCA),
azulene (AZU), trans-stilbene
(STI), acenaphthylene (ACE),
anthracene (ANT), phenan-
threne (PHE), and tetracene
(TET) were chosen as guests to
study the solubility enhance-
Scheme 2. Synthesis of octasubstituted g-CD thioethers: a) 1) PPh3, I2, DMF; 2) Ac2O, pyridine; 3) NaOMe,
MeOH; b) the corresponding thiol compound, triethylamine, DMF.
ments and both stabilities and
stoichiometries of the com-
plexes. It should be noted that
some of the aromatic guests such as STI and ANT can serve
as candidates for photodimerization reactions.
cation, the octakis-iodo derivative was completely acetylat-
ed, subjected to flash chromatography, and deacetylated
again. The target compounds were synthesized by nucleo-
philic displacement reactions in good yields. The reaction
proceeded efficiently, and subsequent selective precipitation
of the desired product in proper organic solvent allowed the
isolation of the octakis-g-CD derivative. The structures of g-
CD derivatives were characterized by NMR spectroscopy
and electrospray ionization mass spectrometry (ESI-MS).
Further purifications of these compounds were achieved by
nanofiltration in water. Except for derivative 7, all other
synthesized neutral and ionic g-CD derivatives showed very
high water solubilities. The structures of the substituents R,
the obtained yields, and the aqueous solubilities of g-CD de-
rivatives 1–8 are summarized in Table 1.
Results and Discussion
Synthesis and Characterization
Hydrophilic thioethers 1–8 at all primary carbon atoms of g-
CD were synthesized from octakis(6-deoxy-6-iodo)-g-CD[13]
as shown in Scheme 2. For the purpose of a rigorous purifi-
Abstract in Chinese:
Solubility Studies
Aqueous solubilities of the eight aromatic guests (shown in
Scheme 1) were measured by UV/Vis spectroscopy as func-
tions of the concentrations of hosts. These solubilities in the
presence of native CDs and g-CD derivatives for a fixed
Chem. Asian J. 2011, 6, 2390 – 2399
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2391

H. M. Wang and G. Wenz
FULL PAPERS
Table 1. Substituents R, yields, and solubilities of the synthesized g-CD
derivatives in g/100 g water at 258C.
tration of dissolved guest versus the total concentration of
host in solution. According to Higuchi and Connors, they
are divided into two main categories, type A and B.[14] Type-
A curves, which indicate the formation of soluble inclusion
complexes, are obtained when the solubility of the guest in-
creases with increasing CD concentration over the entire
concentration range. A linear (AL-type) relationship is dis-
tinguished from positively (AP-type) or negatively (AN-type)
deviating isotherms. Bs-type and BI-type phase-solubility di-
agrams represent the formation of complex with limited sol-
ubility and insoluble complex, respectively.
The obtained phase-solubility diagrams are representa-
tively shown in Figure 1. The solubility of PHE increased
linearly with the increasing concentrations of native CDs
and ionic octasubstituted g-CD derivatives 3, 4, and 6, thus
leading to an AL type of classification, shown in Figure 1a.
On the other hand, curved phase-solubility diagrams were
obtained for the complexes of NCA in a-CD and g-CD de-
rivative 1, shown in Figure 1b. NCA exhibited a BS-type dia-
gram with a-CD in water, possibly due to a limited solubili-
ty of the formed inclusion complex.[14b] The diagram of
NCA with derivative 1 showed AN-type curvature, which
was attributed to electrostatic interaction between the
amino groups of 1 and the carboxyl group of NCA, which
might lead to aggregation of the CD inclusion complex at
higher concentrations.[14b,15] Nevertheless, most complexes in
Table 2 show the linear AL-type phase-solubility diagrams,
which can be interpreted by the formation of soluble 1:1, as
well as 1:2 host–guest complexes.[14a,16] The complex stoichi-
ometry remained an open question.
No.
1
R
Yield [%]
71
Solubility [%]
21
2
3
62
92
40
49
4
41
42
5
6
85
82
22
24
7
8
86
81
0.3
48
host concentration of 6 mm, as well as the intrinsic water sol-
ubilities of the guests, are listed in Table 2. Although the sol-
ubility enhancements caused by native CDs were quite
small, significant enhancements were found for the thioeth-
ers, thereby revealing very strong intermolecular interac-
tions between CD derivatives 1–8 and these aromatic guests
in water. In particular, the solubilities enhancements for
NAP, AZU, and ANT are much higher than for known CD
derivatives like hydroxypropyl-b-CD.[10c] The ionic hosts 1,
3, and 6 generally performed better than the neutral ones 5,
7, and 8. Most systems show a linear increase of solubility
with host concentrations, displayed in bold characters in
Table 2, whereas a few systems show nonlinear solubility en-
hancements, displayed in italic characters.
The examination of so-called phase-solubility diagrams is
widely used to investigate both the complex stoichiometry
and binding constants of host–guest complexes. A phase-sol-
ubility diagram is obtained by plotting the observed concen-
Table 2. Intrinsic water solubilities of guests and equilibrium solubilities
of guests in 6.0 mm solutions of native CDs and g-CD derivatives 1–8.[a]
Host
Guest [mm]
NAP
NCA
AZU
STI
ACE
ANT
PHE
TET
water
a-CD
b-CD
g-CD
1
2
3
4
189
205
234
196
2440
479
1590
481
258
1080
199
515
322
913
1460
431
6180
11400
3570
9850
2180
3420
863
1.4
26
65
0.6
5.0
17
1.5
67
13
23
13
14
22
2.0
50
67
85
0.4
2.4
4.6
1.6
44
0.6
1.9
45
5.9
102
60
100
64
56
0.2
0.2
0.3
0.5
8.9
5.5
6.8
5.0
3.2
6.2
0.2
3.2
105
103
898
729
1703
693
278
1280
95
31
588
320
841
149
390
670
20
15
13
6.9
7.4
12
0.8
14
5
6
132
19
62
7[b]
8
1200
353
162
[a] Bold characters: AL-type linear phase-solubility diagram; italic char-
acters: nonlinear phase-solubility diagram. [b] The concentration of CD
derivative 7 was only 1.5 mm due to its low water solubility.
Figure 1. Phase-solubility diagrams for a) PHE and b) NCA in aqueous
solution in the presence of native CDs and g-CD derivatives 6, 3, 4, and
1.
2392
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2390 – 2399

Polycyclic Aromatics
Table 4. Absorption (lab) and emission (lem) maxima [nm], and Stokes shift values
(Dl) [nm] of aromatic guests in water in the presence of b-CD derivative b-1 and g-
CD derivative 1 at 298 K.
Complex Stoichiometries
The slope m of the phase-solubility diagram, equiv-
alent to the average amount of guest per host,
called occupancy, gave some rough idea about the
stoichiometry of a complex. These occupancies of
g-CD cavities by the aromatic guests are listed in
Table 3. Initial slopes m were taken for the curved
phase-solubility diagrams.
Occupancy values of more than 100% were
found for NCA included in CD derivatives 1, 2, and
4, thereby providing strong evidence for the exis-
tence of complexes with 1:2 stoichiometries. An oc-
cupancy of less than 100% found in the other cases
does not necessarily imply a 1:1 stoichiometry, be-
Guest
b-1
lem
1
lem
DDl[a]
lab
Dl
lab
Dl
87
72
71
77
124
74
NAP
NCA
AZU
STI
ACE
ANT
PHE
TET
300
343
365
310
354
361
357
367
325
374
377
356
437
406
368
424
25
31
12
46
83
45
11
57
337
340
353
351
284
294
290
360
424
412
424
428
408
368
367
432
62
41
59
31
41
29
66
15
77
72
[a] DDl is equal to the excess Stokes shift values of guests complexed in g-CD deriva-
tives 1 compared to those complexed in b-1.
cause cooperativity of inclusion has to be taken into ac-
count. If there is a strong interaction between two included
guests, a mixture of empty CDs and 1:2 CD complexes will
prevail. It should be also noted that several 1:2 complexes
with g-CD or g-CD derivatives are already known.[11b,c,k,17]
(15 nm) was significantly less than that of the other guests.
Therefore 1:2 CD–guest complexes are less likely to have
been formed in this case.
Typical absorption and emission spectra of STI in the
presence of CD derivatives in aqueous solution are shown
as examples in Figure 2. A small Stokes shift (Dl=
46 nm) was found for STI complexed in b-1, where-
Table 3. Occupancies [%, equal to slope m] of aromatic guests within the g-CD cavi-
ty.[a,b]
as a large Stokes shift (Dl=77 nm) was observed
for STI in g-CD derivative 1. Both emission bands
were narrow and monomodal, thus demonstrating
that only one species of stilbene complex existed in
each case, namely, the 1:1 and the 1:2 complexes,
respectively. This finding was in contrast to the
fluorescence spectra obtained for the complex of
naphthalene in native b-CD, for which a broad bi-
modal fluorescence was found.[18] The formation of
a 2:2 complex was discussed for the NAP-b-CD
system. This model appeared implausible for the
Host
Guest
NAP
NCA
AZU
STI
ACE
ANT
PHE
TET
g-CD
n.d.
38
3.9
31
7.7
12
18
5.4
165
167
55
123
30
52
34
15
0.5
9.7
6.5
16
2.9
6.9
12
0.03
1.2
0.2
0.4
0.2
0.3
0.4
0.2
0.9
n.d.
13
11
28
11
3.4
21
n.d.
2.5
0.02
0.7
0.2
0.2
0.1
0.1
0.2
0.06
0.2
0.09
1.5
1.1
1.7
1.2
1.9
2.3
1.2
1.1
n.d.
0.1
0.09
0.1
0.08
0.05
0.1
n.d.
0.05
1
2
3
4
5
6
7
8
30
10
1.5
5.7
[a] For the non-AL-type phase-solubility diagrams, initial slopes m are taken. Values
are given in italics. [b] n.d.=solubility enhancement of guest was not detectable.
To gain further insight into the host–guest stoichiometries
of the complexes, fluorescence spectroscopy was employed,
because complexation of two chromophores within a CD
cavity is known to cause excimer formation, thereby leading
to large Stokes shifts. Excimer formation was already ob-
served for CD complexes of NAP,[18] STI,[19] ANT,[20] and
pyrene.[21]
The fluorescence spectra of the inclusion compounds of
the aromatic guests in g-CD derivative 1 showed a single
emission band in each case with pronounced Stokes shifts of
Dl=71 to 124 nm (these are listed in Table 4, and detailed
spectra are collected in the Supporting Information). For
comparison, the corresponding heptakis[6-deoxy-6-(2-ami-
noethylsulfanyl)]-b-CD (b-1),[2c,9] which only accommodates
one single guest, showed much smaller Stokes shifts of Dl=
26 to 83 nm, typical for dilute solutions of the fluorophores
in organic solvents. Consequently, the large extra Stokes
shifts DDl, listed in Table 4, strongly support our hypothesis
of two aromatic guests being included in the large cavity of
Figure 2. Absorption (dashed line) and emission (solid line) spectra of
the g-CD derivative 1. Only the DDl value for guest TET
the inclusion system of a) STI with b-1 and b) CD 1 in aqueous solution.
Chem. Asian J. 2011, 6, 2390 – 2399
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2393

H. M. Wang and G. Wenz
FULL PAPERS
Table 5. In-plane areas A and maximum cross-sectional areas Amax (in ꢂ2) in direction
of the long axis and short axis of the molecule of the aromatic guest molecules.
complexes of aromatic guests in g-CD derivatives
described here, because narrow excimer emission
bands were already observed for very low concen-
trations of g-CD derivative 1. For a further clarifi-
cation of the stoichiometries of complexes of g-CD
derivatives 1–8, molecular packing was investigated
by molecular modeling studies.
Guest
NAP
NCA
AZU
STI
ACE
ANT
PHE
TET
Amax long axis
Amax short axis
A in plane
31.4
20.0
55.0
62.0
23.4
65.0
40.2
22.7
55.5
82.3
19.9
78.4
31.4
28.5
61.7
64.3
21.6
62.8
64.3
24.1
72.0
100.0
21.6
90.3
Molecular Modeling
Although the modification of g-CD on the primary rim with
side arms changes the height of g-CD, the internal diameter
d of the g-CD should remain unaffected. According to this
view, g-CD was selected as a simple model host throughout
the theoretical calculations. Quantum mechanical calcula-
tions were employed to obtain the inner surface (electron
density map) of the g-CD cavity. The cross-sectional area A
of the g-CD cavity was obtained by analysis of the electron
density map of g-CD with the program MolShape,[22] which
was already useful for prediction of stoichiometries of poly-
rotaxanes.[2b] Thus, the corresponding profile of the inner
width d of the g-CD cavity along the C8 axis was obtained.
The maximum inner diameter (cross-sectional area) at the
secondary rim of g-CD was determined to be dmax =10.3 ꢂ
Scheme 3. Schematic drawing depicting the top view of 1:2 g-CD com-
plexes for three possible orientations: a) long axis, b) short axis, and
c) both axes perpendicular to the C8 axis of g-CD.
sidered (i.e., the parallel ›› and antiparallel ›fl orienta-
tions). Given that dispersion forces are known to play an
important role in the energetics of aromatic dimers, it is im-
perative that the calculation of the energies has to be made
at the level of second-order Møller–Plesset perturbation
theory (MP2).[25] The aromatic dimers were fully optimized
at the MP2/6-31G* level without any symmetry restriction.
The resulting interaction energies DE, the center of mass
distances d, and the maximum cross-sectional areas Amax in
the direction of the short axis are collected in Table 6. The
interaction energy DE became more negative with increas-
ing size of the molecule, and was in good accordance with
published data.[25,26] Remarkable differences in DE values of
the aromatic dimers, which possess two stable orientations,
were clearly observed from Table 6. For instance, the DE
value of NCA dimer with antiparallel orientation was found
to be significantly more negative than that of the dimer with
parallel orientation. This result can be attributed to the ad-
(Amax =83.5 ꢂ2). The minimum internal diameter dmin
=
8.0 ꢂ (Amin =50.6 ꢂ2), situated approximately at hydrogen
H-5 of the g-CD cavity was found, which was in good ac-
cordance with the results of both experimental measure-
ments and other theoretical calculations.[1c,22,23] Similarly, the
maximum cross-sectional areas Amax of the aromatic guests
were calculated, both in direction of the short and the long
axis of the molecule, and also the areas A in the plane of
the molecules, summarized in Table 5. The dimeric aromatic
guest can fit principally in three possible orientations, with
the long axis, the short axis, and both axes perpendicular to
the C8 axis in the g-CD cavity, depicted in Scheme 3. From
these calculations it became clear that all guests can be ac-
commodated in the complete g-CD cavity as dimers with
their short axis oriented perpendicular to the C8 axis
(Scheme 3b), because twice the corresponding cross-section-
al area Amax of the guest is less or equal to the minimum
cross-sectional area Amin =50.6 ꢂ2 of the host. The predomi-
nance of this orientation is in accordance with other experi-
mental results.[11c–e,k] The other orientations are not possible
because of geometrical constraints. The smaller dimensions
of the b-CD cavity, Amin =34.6 ꢂ2 and Amax =60.4 ꢂ2 explain
why the NAP dimer can only protrude partially in it, which
leads to the formation of 2:2 complexes, necessary for a
complete coverage of the dimer.[18]
Table 6. The intermolecular interaction energies (DE) and the distances
d of the mass centers of the two guests within the dimer, and the calculat-
ed maximal cross-sectional areas (Amax
) of the optimized aromatic
dimers.
Dimer[a]
DE [kJmolꢀ1
]
d [ꢂ]
Amax [ꢂ2]
NAP
ꢀ17.8
ꢀ18.0
ꢀ21.8
ꢀ34.6
ꢀ22.3
ꢀ20.5
ꢀ29.2
ꢀ31.0
ꢀ27.6
ꢀ21.9
3.64
3.69
3.60
3.29
3.91
3.54
3.18
3.65
3.86
3.85
44.8
52.3
52.0
51.0
57.1
53.1
52.0
49.6
56.8
66.7
NCA (dimer ›fl)
AZU (dimer ››)
AZU (dimer ›fl)
STI
ACE (dimer ››)
ACE (dimer ›fl)
ANT
Quantum mechanical calculations of the structures and in-
teraction energies DE of the aromatic dimers were per-
formed using the Gaussian 03 software package.[24] Former
experimental and theoretical investigations on the aromatic
dimers indicated that the dimers with the largest overlap-
ping areas generally would have the most negative values of
intermolecular interaction energy. Accordingly, in this work,
two starting orientations of the aromatic dimers were con-
PHE (dimer ››)
PHE (dimer ›fl)
[a] For the NCA dimer ››, a T-shaped orientation was obtained after op-
timization.
2394
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2390 – 2399

Polycyclic Aromatics
ditional dipole–dipole interaction between the two mono-
mers in the antiparallel orientation. Similar phenomena also
occurred for the aromatic dimers of AZU and ACE. The
corresponding distances d between two monomers in these
parallel-configured dimers were found to be remarkably
larger than those of the monomers with antiparallel orienta-
tion. Contrarily, the interaction energy of the parallel-orient-
ed dimer of PHE was more negative than the one of the an-
tiparallel dimer. It appeared reasonable that the relatively
poor overlap between two PHE molecules in the antiparal-
lel dimer caused this result.[27]
It has to be noted that the cross-sectional areas Amax de-
rived from the calculated structures of the dimers are larger
than twice of the calculated cross-sectional area of the cor-
responding monomer. But these more realistic Amax values
from Table 6 are still small enough to fit in the g-CD cavity.
The packing of AZU dimer in the g-CD cavity is provided
as an example in Figure 3. This dimer fits well into the
cavity of g-CD without steric hindrance. As can be seen,
both orientations within the dimer, parallel (Figure 3b) and
antiparallel (Figure 3c), were perfectly accommodated into
the cavity of g-CD. The antiparallel orientation of AZU ap-
pears to be more probable because of its more negative DE
value. Consequently, taking together all facts such as the
very good space filling of the g-CD cavity, the sufficiently
large negative values of DE, and the observed large Stokes
shifts, the formation of 1:2 host–guest complexes became
evident. After determination of the stoichiometry, binding
constants and binding free energies could be finally calculat-
ed.
Binding Free Energy
Binding constants K and binding free energies DG were cal-
culated from the slope m of the solubility isotherm and the
solubilities of the free guests in water according to Equa-
tions (7) and (8), respectively (see the Experimental Section
below). The calculated DG values of the aromatic guests
with g-CD or g-CD derivatives are summarized in Table 7.
As discussed above, as TET is less likely to form an 1:2 in-
clusion complex with g-CD or g-CD derivatives, it was ex-
cluded from the discussion. Pronounced differences between
the DG values were observed for the various guests. Guests
AZU, ANT, and PHE showed more negative DG values
(around ꢀ55 kJmolꢀ1) than NAP and NCA (around
ꢀ35 kJmolꢀ1), which was in good correlation with the more
negative interaction energies DE of AZU, ANT, and PHE
dimers (around ꢀ31 kJmolꢀ1) relative to NAP and NCA
(around ꢀ19 kJmolꢀ1). This means that the interaction
energy of the dimer provides a major contribution to bind-
ing free energy. Despite its strongly negative DE value,
ACE only shows a moderate DG value, which might be due
to its large cross-sectional area along its short axis, which
might lead to a very tight fit within the g-CD cavity. In our
previous work, we found that a loose fit of a guest in a CD
cavity leads to a higher binding constant than a tight fit, pos-
sibly due to entropic effects.[22] These entropic effects would
also explain why the slim guest STI shows a strongly nega-
tive DG value, whereas its DE value is only moderate.
Conclusion
In the present work, we have synthesized and characterized
a series of octasubstituted g-CD derivatives with various
neutral or ionic side arms bound through thioether linkages
to the CD scaffold. It was found that these g-CD derivatives
showed much higher solubilization abilities for aromatic
guests than native CDs or other frequently used derivatives
such as hydroxypropyl-b-CD. The existence of 1:2 host–
guest complexes was confirmed from the phase-solubility
data, the excimer emissions, as well as theoretical investiga-
tions. The DG values of the CD–guest complexes obtained
from the solubility data are strongly negative because the in-
teraction energy between the included guests makes the
Figure 3. Schematic drawing depicting the AZU dimer in g-CD: a) top
view, b) side view ›› orientation, and c) side view ›fl orientation.
Chem. Asian J. 2011, 6, 2390 – 2399
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2395

H. M. Wang and G. Wenz
FULL PAPERS
Table 7. Binding free energies (DG) [kJmolꢀ1] for the complexation of aromatic
guests in g-CD and g-CD derivatives at 298 K.
added. The solution was concentrated in vacuo to half volume
after stirring at 708C for 24 h. NaOMe in MeOH (3m, 160 mL)
was then added under cooling and stirred for 30 min at RT.
The thick reaction mixture was poured into methanol (2.5 L)
under vigorous stirring to form a brownish precipitate, which
was filtered, washed afterwards with MeOH (0.5 L), and dried.
The brown solid was dissolved in dry pyridine (130 mL). Ac2O
(130 mL) was added, and the resulting solution was stirred for
2 d at room temperature. Methanol (70 mL) was added and
stirred for 1 h to quench the reaction. Column chromatography
on silica gel and ethyl acetate (EtOAc)/n-hexane (4:1) as an
eluent gave a pale yellow solid (Rf =0.2–0.3). The raw product
was then suspended in anhydrous MeOH (200 mL) that con-
tained NaOMe (0.50 g) for 4 d. Compound 9 was recovered as
Host
Guest[a]
STI
NAP
NCA
AZU
ACE
ANT
PHE
g-CD
-
ꢀ31.0
–
ꢀ51.6
ꢀ59.3
ꢀ58.3
ꢀ60.7
ꢀ56.3
ꢀ58.5
ꢀ59.9
ꢀ54.6
ꢀ58.0
ꢀ48.9
ꢀ58.5
ꢀ54.1
ꢀ56.0
ꢀ54.3
ꢀ54.8
ꢀ55.9
ꢀ53.4
ꢀ57.9
–
ꢀ50.3
ꢀ59.1
ꢀ56.4
ꢀ56.1
ꢀ54.4
ꢀ54.6
ꢀ56.0
ꢀ52.0
ꢀ56.4
ꢀ52.1
ꢀ59.2
ꢀ58.4
ꢀ59.6
ꢀ58.7
ꢀ57.7
ꢀ60.4
ꢀ53.0
ꢀ58.6
1
2
3
4
5
6
7
8
ꢀ38.9
ꢀ32.8
ꢀ38.3
ꢀ34.5
ꢀ35.8
ꢀ36.8
ꢀ38.2
ꢀ35.2
ꢀ40.9
ꢀ40.7
ꢀ43.2
ꢀ40.6
ꢀ37.6
ꢀ42.4
–
ꢀ43.9
ꢀ37.4
ꢀ41.0
ꢀ35.5
ꢀ37.3
ꢀ35.9
ꢀ33.7
ꢀ36.8
a
white powder after suction filtration. Yield: 28.54 g
[a] DG values given in italics were calculated on the basis of the initial slope m of the
phase-solubility diagrams.
1
(13.1 mmol, 66%); H NMR (400 MHz, [D6]DMSO): d=5.96–
5.97 (d, J=2.8 Hz, 8H; OH-2), 5.95 (s, 8H; OH-3), 5.02–5.03
(d, J=3.6 Hz, 8H; H-1), 3.79–3.82 (d, J=9.2 Hz, 8H; H-3),
3.57–3.67 (m, 16H; H-6a,b), 3.44–3.55 (m, 16H; H-4,5), 3.25–3.30 ppm (t,
J=9.0 Hz, 8H; H-2); 13C NMR (100 MHz, [D6]DMSO): d=9.25, 71.08,
71.80, 72.37, 85.24, 101.93 ppm.
main contribution to the driving force of complex forma-
tion.
These g-CD derivatives are good candidates for molecular
reaction vessels for photodimerization reactions of native
polycyclic aromatic molecules in aqueous solution. The sig-
nificantly enhanced solubilities of the aromatic guests in
water would allow photodimerization of these guests under
homogenous conditions for the first time. The kinetics as
well as the stereo- and regioselectivities[28] of the photodime-
rizations of these guests mediated by the g-CD derivatives
will be investigated in our future works.
General Procedure for the Synthesis of Neutral Octakis-6-deoxy-g-CD
A
procedure identical to that described previously was utilized
(Scheme 2).[2a] Triethylamine (3.98 mL, 28.57 mmol) and the correspond-
ing thiol compound (28.57 mmol) were added with stirring to a solution
of 9 (3.11 g, 1.43 mmol) in DMF (15 mL). After stirring for 3 d at 608C
under N2, the crude product was diluted with sufficient amount of appro-
priate organic solvents. The resulting precipitate was filtered off, washed,
dried in vacuo, and then purified by ultrafiltration.
General Procedure for the Synthesis of Ionic Octakis-6-deoxy-g-CD
Triethylamine (3.98 mL, 28.57 mmol) and the corresponding methyl ester
of the thiol compound (28.57 mmol) were added to a solution of com-
pound 9 (3.11 g, 1.43 mmol) in DMF (15 mL). After stirring for 3 d at
608C under N2 atmosphere, the reaction mixture was diluted with suffi-
cient amount of appropriate organic solvents. The resulting precipitate
was collected, then stirred in NaOH (1m) for 18 h, and then further puri-
fied by ultrafiltration.
Experimental Section
General
Unless otherwise stated, all chemicals were used as received. Corre-
sponding thiol compounds for the synthesis of g-CD derivatives 1–7 were
purchased from Sigma Aldrich. The thiol compound for the synthesis of
8 was synthesized and purified according to a published procedure.[29]
Octakis[6-deoxy-6-(2-aminoethylsulfanyl)]-g-CD (1)
Compound 1 was synthesized according to the general procedure. The
crude product was precipitated in acetone, dissolved in water, ultrafil-
tered, and lyophilized. Yield: 1.80 g (71%), white solid; 1H NMR
(400 MHz, D2O, 25 8C): d=5.17–5.18 (d, J=3.6 Hz, 8H; H-1), 3.88–3.97
(m, 16H; H-3,5), 3.65–3.69 (dd, J=10.0 Hz, 8H; H-2), 3.52–3.57 (t, J=
9.4 Hz, 8H; H-4), 2.91–3.28 (m, 16H; H-6a,6b), 2.80–2.86 ppm (dd, J=
20.0 Hz, 32H; H-7,8); 13C NMR (100 MHz, D2O, 25 8C): d=35.33, 39.97,
64.56, 71.65, 72.70, 83.32, 100.01 ppm; MS (3.80 kV, ESI, water): m/z
Solvents were dried as follows: DMF over P2O5 and distilled; pyridine
over KOH; and methanol distilled from iodine and magnesium turnings
immediately before use.[30] a-, b-, and g-CD were dried in vacuo at 1008C
overnight. g-CD derivatives were purified by nanofiltration (molecular
weight cutoff: 1000 Da).
Teflon syringe filters (0.22 mm) from Roth, Karlsruhe, Germany were
used to remove insoluble material before UV/Vis and fluorescence meas-
urements. Thin-layer chromatography (TLC) was performed on alumi-
num sheets coated with silica gel (F254) and developed by charring with
5% H2SO4 in ethanol. Electrospray ionization mass spectrometric (ESI-
MS) analysis was performed with an ESI single-quadrupole mass spec-
trometer (micromass ZQ 4000, Waters) from solutions in water. The fol-
lowing settings were used: capillary voltage 3.80 kV, cone voltage 20 V,
extractor voltage 5 V. UV/Vis spectra of aqueous samples were per-
formed with a Perkin–Elmer Lambda 2 spectrometer (l: 200–600 nm),
using quartz cells with a 1 cm or 1 mm optical path at 298 K. Fluores-
cence spectra were recorded with a JASCO spectrophotometer using
quartz cells of 10.0 mm path at 298 K. All NMR spectra were recorded
with a Bruker 400 MHz NMR spectrometer at 298 K using the solvent
peaks as internal references. Coupling constants (J) were measured in
Hz. The signals are described as follows: s (singlet), d (doublet), t (trip-
let), and m (multiplet).
(%): 1769.84 (20) [M+H]+, 884.40 (79) [M+2H]2+
[M+3H]3+
, 589.57 (100)
.
Octakis[6-deoxy-6-(2-sulfanyl acetic acid)]-g-CD (2)
Compound 2 was synthesized according to the general procedure. The
crude product was precipitated in ethanol, stirred in 1m NaOH for 18 h,
ultrafiltered, and lyophilized. Yield: 1.82 g (62%), white solid; 1H NMR
(400 MHz, D2O, 25 8C): d=5.13–5.14 (d, J=3.6 Hz, 8H; H-1), 4.01–4.05
(m, 8H; H-5), 3.87–3.92 (t, J=9.6 Hz, 8H; H-3), 3.66–3.70 (t, J=9.4 Hz,
8H; H-2), 3.59–3.62 (dd, J=10.0 Hz, 8H; H-4), 3.30–3.38 (m, 16H; H-7),
2.90–3.12 ppm (m, 16H; H-6a,6b); 13C NMR (100 MHz, D2O, 25 8C): d=
33.73, 38.60, 71.37, 72.31, 72.61, 82.45, 101.19, 177.71 ppm; MS (3.80 kV,
ESI, water): m/z (%): 1912.60 (25) [M+Na]+, 967.84 (100) [M+2Na]2+
.
Octakis[6-deoxy-6-(2-sulfanylethanesulfonic acid)]-g-CD (3)
Octakis(6-deoxyiodo)-g-CD (9)
Compound
9 (3.11 g, 1.43 mmol), sodium 2-mercaptoethanesulfonate
As shown in Scheme 2, Ph3P (104.92 g, 400 mmol) was dissolved in anhy-
drous DMF (320 mL). Under an N2 atmosphere, I2 (104.06 g, 410 mmol)
was added over 15 min at 08C. Dry g-CD (25.94 g, 20 mmol) was then
(4.60 g, 28.00 mmol), and triethylamine (3.98 mL, 28.57 mmol) were dis-
solved in DMSO (25 mL). After stirring for 3 d at 608C under N2, the
product was precipitated in acetone, dissolved in water, ultrafiltered, and
2396
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2390 – 2399

Polycyclic Aromatics
lyophilized. Yield: 3.25 g (92%), white solid; 1H NMR (400 MHz, D2O,
258C): d=5.14–5.15 (d, J=3.6 Hz, 8H; H-1), 4.00–4.05 (m, 8H; H-5),
3.88–3.93 (t, J=9.6 Hz, 8H; H-3), 3.58–3.62 (m, 16H; H-2,4), 3.11–3.21
(m, 24H; H-6a,7), 2.91–2.99 ppm (m, 24H; H-6b,8); 13C NMR (100 MHz,
D2O, 25 8C): d=27.27, 33.27, 51.37, 71.32, 72.26, 72.68, 100.01,
100.96 ppm; MS (3.80 kV, ESI, water): m/z (%): 1257.66 (100)
sured in 1,4-dioxane and dichloromethane, respectively. The observed e
values of the aromatic guests at their absorption maxima (l) are summar-
ized in Table 8.
Table 8. Molar extinction coefficients (e) [Lmolꢀ1 cmꢀ1] of guests at the
absorption maxima (l) [nm].
[M+2Na]2+
.
Guest
NAP NCA STI
AZU ACE
ANT
PHE
293
TET
277
Octakis[6-deoxy-6-(2-sulfanylpropanoic acid)]-g-CD (4)
l
e
275
280
310
339
322
356
7350
Compound 4 was synthesized according to the general procedure. The
crude product was precipitated in 2-propanol, stirred in 1m NaOH for
18 h, ultrafiltered, and lyophilized. Yield: 1.27 g (41%), white solid;
1H NMR (400 MHz, D2O, 25 8C): d=5.21–5.22 (m, 8H; H-1), 3.98–4.09
(m, 16H; H-3,5), 3.50–3.76 (m, 24H; H-2,4,7), 3.11–3.22 (m, 8H; H-6a),
2.96–2.99 (m, 8H; H-6b), 1.43–1.45 ppm (d, J=7.2 Hz, 24H; H-8). MS
(3.80 kV, ESI, water): m/z (%): 2025.30 (65) [M+Na]+, 1024.57 (100)
5590
6950
18570 4370
9090
11710 100700
Solubility measurements of the guests in the presence of native CDs and
g-CD derivatives in water were carried out according to the method pro-
posed by Higuchi and Connors.[14] In glass vials that contained excess
amounts of guest molecules, aqueous solutions of native CDs or g-CD
derivatives (concentration of CD: 1.0, 2.0, 3.0, 4.0, and 6.0 mm) were
added. The vials were sealed, protected from light, and magnetically
stirred at room temperature. After 72 h, the solid residues were removed
by filtration with syringe filter. According to Lambert–Beerꢃs law, the
concentrations of guests in pure water and in CDs solutions were deter-
mined from UV/Vis extinctions at the absorption maxima.
[M+2Na]2+
.
Octakis[6-deoxy-6-(3-sulfanylpropane-1,2-diol)]-g-CD (5)
Compound 5 was synthesized according to the general procedure. The
crude product was precipitated in ethanol, dissolved in water, ultrafil-
tered, and lyophilized. Yield: 2.44 g (85%), white solid; 1H NMR
(400 MHz, D2O, 25 8C): d=5.18–5.19 (d, J=3.2 Hz, 8H; H-1), 4.00–4.05
(t, J=8.4 Hz, 8H; H-5), 3.90–3.95 (m, 16H; H-3,8), 3.57–3.72 (m, 32H;
H-2,4,7), 3.25–3.28 (d, J=12.8 Hz, 8H; H-6a), 3.00–3.03 (m, 8H; H-6b),
2.74–2.92 ppm (m, 16H; H-9); 13C NMR (100 MHz, D2O, 25 8C): d=
34.00, 35.93, 64.54, 71.04, 71.45, 72.26, 72.70, 83.32, 101.33 ppm; MS
(3.80 kV, ESI, water): m/z (%): 2039.63 (29) [M+Na]+, 1030.89 (100)
Determination of Binding Free Energy
The increase in water solubility of a guest molecule is due to the intermo-
lecular interactions between the guest molecule and CD to form soluble
complexes.[5b,14b,32] The binding constants (K1:n) for the formation of a 1:n
inclusion complex between CD and guest molecule can be expressed as
Equation (1):
[M+2Na]2+
.
½HGꢁ
Octakis[6-deoxy-6-(3-sulfanylpropanoic acid)]-g-CD (6)
ð1Þ
K1:n
¼
n
½Hꢁ ꢂ ½Gꢁ
Compound 6 was synthesized according to the general procedure. The
crude product was precipitated in ethanol, stirred in 1m NaOH for 18 h,
ultrafiltered, and lyophilized. Yield: 2.55 g (82%), white solid; 1H NMR
(400 MHz, D2O, 25 8C): d=5.15–5.14 (d, J=3.2 Hz, 8H; H-1), 4.01–4.04
(t, J=8.4 Hz, 8H; H-5), 3.89–3.93 (t, J=9.4 Hz, 8H; H-3), 3.58–3.64 (m,
16H; H-2,4), 2.93–3.11 (m, 16H; H-6a,b), 2.80–2.84 (t, J=7.4 Hz, 16H;
H-7), 2.43–2.47 ppm (m, 16H; H-8); 13C NMR (100 MHz, D2O, 25 8C):
d=29.54, 33.34, 37.75, 64.56, 71.29, 73.01, 83.56, 100.01, 180.73 ppm; MS
(3.80 kV, ESI, water): m/z (%): 2023.63 (20) [M+Na]+, 1023.89 (100)
in which n is the guest–host ratio, and [H], [G], and [HG] are the con-
centrations of the free CD, the free guest, and the corresponding inclu-
sion complex, respectively.
Assuming that the concentration of the free guest in CD solution is equal
to that in pure water [G]0, the following equations can be obtained
[Eqs. (2), (3), and (4)]:[33]
½Gꢁ ¼ ½Gꢁ0
ð2Þ
ð3Þ
ð4Þ
[M+2Na]2+
.
½HGꢁ ¼ ð1=nÞð½Gꢁtꢀ½Gꢁ0Þ
½Hꢁ ¼ ½Hꢁtꢀ½HGꢁ
Octakis[6-deoxy-6-(2-hydroxyethylsulfanyl)]-g-CD (7)
Compound 7 was synthesized according to the general procedure. The re-
action mixture was poured into ethanol. The formed precipitate was col-
lected by suction filtration, recrystallized in cold water, and dried to give
in which [G]0 is the equilibrium solubility of guest in pure water, and [H]t
and [G]t correspond to the total concentrations of CD and guest, respec-
tively.
7
(yield: 2.19 g, 86%) as colorless needles. 1H NMR (400 MHz,
[D6]DMSO, 258C): d=5.91 (s, 16H; OH-2,3), 4.92–4.93 (d, J=3.6 Hz,
8H; H-1), 4.70–4.72 (t, J=5.4 Hz, 8H; OH-9), 3.72–3.76 (m, 8H; H-5),
3.50–3.59 (m, 24H; H-2,3,4), 3.32–3.41 (m, 16H; H-7), 2.79–3.09 (m,
16H; H-6a,b), 2.62–2.67 ppm (m, 16H; H-8); MS (3.80 kV, ESI, water):
m/z (%): 1799.52 (65) [M+Na]+, 911.37 (100) [M+2Na]2+, 1775.88 (100)
[MꢀH]ꢀ.
Substituting Equations (2–4) into Equation (1) affords Equation (5):
½Gꢁt ꢀ ½Gꢁ0
K1:n
¼
ð5Þ
n
½Gꢁ0 ꢂ ð½Hꢁt ꢀ ð½Gꢁt ꢀ ½Gꢁ0Þ=nÞ
The slope (m) of the dependence line is represented by Equation (6) as
follows:
Octakis[6-deoxy-6-(10-sulfanyl-2,5,8-trioxodecane)]-g-CD (8)
½Gꢁt ꢀ ½Gꢁ0
Compound 8 was synthesized according to the general procedure. The
crude product was precipitated in diethyl ether, ultrafiltered, and lyophi-
lized. Yield: 3.02 g (81%), white solid; 1H NMR (400 MHz, D2O, 25 8C):
d=5.03–5.04 (d, J=3.6 Hz, 8H; H-1), 3.71–3.80 (m, 16H; H-3,5), 3.62–
3.66 (m, 16H; H-2,4), 3.40–3.59 (m, 80H; H-10), 3.29 (s, 24H; H-9),
2.80–3.25 ppm (m, 32H; H-6,7,8); MS (3.80 kV, ESI, water): m/z (%):
1297.63 (35) [M+2H]2+, 2593.48 (100) [MꢀH]ꢀ.
m ¼
ð6Þ
½Hꢁt
A plot of [G]t against [H]t yields a straight line. Thus, the binding con-
stant between CD and guest can be calculated from the phase diagrams
using Equation (7):
m
K1:n
¼
ð7Þ
n
½Gꢁ0 ꢂ ðn ꢀ mÞ
Phase-Solubility Investigations
Since the effective polarity of the CD cavity was proven to be similar to
that of ethanol,[10c,31] the molar extinction coefficients (e) of NAP, NCA,
AZU, ACE, ANT, and PHE were measured in ethanol. However, due to
their low solubilities in ethanol, the e values of STI and TET were mea-
The binding free energy (DG) is calculated according to Equation (8):
DG ¼ ꢀRT ln K
ð8Þ
Chem. Asian J. 2011, 6, 2390 – 2399
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2397

H. M. Wang and G. Wenz
FULL PAPERS
Molecular Modeling
[6] T. Loftsson, A. Magnusdottir, M. Masson, J. F. Sigurjonsdottir, J.
Pharm. Sci. 2002, 91, 2307–2316.
All quantum mechanical calculations were carried out with the Gaussi-
an 03 software package.[24] The initial geometries of the aromatic guests
were built up from their respective crystal structures[34] and then fully op-
timized without any symmetrical restrictions at the MP2/6-31G* level.
The input structures of b- and g-CD were taken from existing crystallo-
graphic data[23,35] and fully optimized by the PM3 method.[36] The glycosi-
dic oxygen atoms of CDs were placed onto the xy plane, and the center
of the glycosidic oxygen atoms was designated as the origin of the Carte-
sian coordinate system. The secondary OH rim of CDs was placed point-
ing toward the positive z axis. The rendering was performed with VMD
1.8.7.[37]
[7] a) G. Wenz, T. Hofler, Carbohydr. Res. 1999, 322, 153–165; b) J.
Pitha, J. Milecki, H. Fales, L. Pannell, K. Uekama, Int. J. Pharm.
1986, 29, 73–82.
[8] V. Zia, R. A. Rajewski, V. J. Stella, Pharm. Res. 2001, 18, 667–673.
[9] K. Kano, Y. Ishida, K. Kitagawa, M. Yasuda, M. Watanabe, Chem.
Asian J. 2007, 2, 1305–1313.
[10] a) M. Lubomska, P. Gierycz, M. Rogalski, Fluid Phase Equilib. 2005,
238, 39–44; b) X. Wang, M. L. Brusseau, Environ. Sci. Technol.
1995, 29, 2346–2351; c) X. Wang, M. L. Brusseau, Environ. Sci.
Technol. 1993, 27, 2821–2825.
[11] a) K. Takahashi, Chem. Rev. 1998, 98, 2013–2033; b) X. Wu, L. Lei,
L. Wu, G. Liao, L. Luo, X. Shan, L. Zhang, C. Tung, Tetrahedron
2007, 63, 3133–3137; c) L. Luo, G. H. Liao, X. L. Wu, L. Lei, C. H.
Tung, L. Z. Wu, J. Org. Chem. 2009, 74, 3506–3515; d) C. Yang, A.
Nakamura, T. Wada, Y. Inoue, Org. Lett. 2006, 8, 3005–3008; e) A.
Nakamura, Y. Inoue, J. Am. Chem. Soc. 2005, 127, 5338–5339; f) S.
Karthikeyan, V. Ramamurthy, Tetrahedron Lett. 2005, 46, 4495–
4498; g) N. Haga, H. Takayanagi, K. Tokumaru, J. Org. Chem. 1997,
62, 3734–3743; h) F. D. Lewis, S. V. Barancyk, E. L. Burch, J. Phys.
Chem. 1992, 96, 2548–2553; i) K. Wei, R. Livingston, Photochem.
Photobiol. 1967, 6, 229–232; j) J. Reichwagen, H. Hopf, A. Del
Guerzo, J. Desvergne, H. Bouas-Laurent, Org. Lett. 2004, 6, 1899–
1902; k) W. Herrmann, S. Wehrle, G. Wenz, Chem. Commun. 1997,
1709–1710.
[12] A. Nakamura, Y. Inoue, J. Am. Chem. Soc. 2003, 125, 966–972.
[13] P. R. Ashton, R. Koniger, J. F. Stoddart, D. Alker, V. D. Harding, J.
Org. Chem. 1996, 61, 903–908.
[14] a) R. Singh, H. H. Tonnesen, S. B. Vogensen, T. Loftsson, M.
Masson, J. Inclusion Phenom. Macrocyclic Chem. 2010, 66, 335–
348; b) T. Higuchi, K. Connors, Adv. Anal. Chem. Instrum. 1965, 4,
117–212; c) T. Higuchi, H. Kristiansen, J. Pharm. Sci. 1970, 59,
1601–1608.
Cross-Sectional Area
The electron density function was transformed into a discrete cubic elec-
tron density by using the program Cubegen from Gaussian 03.[24] The
cross-sectional diameters and the cross-sectional areas of CD and the
guest molecules were then calculated from the cubic electron densities by
the program MolShape.[22] An electron density cutoff value of 0.002 au
was chosen to define the surface of the molecule.[22]
Aromatic Dimer
The aromatic dimers were built up by combining two optimized mono-
mers together, and then used as initial guesses for full geometry optimi-
zations of the gas-phase dimers at the MP2/6-31G* level of theory. The
planes of two monomers with a center-to-center distance of 3.5 ꢂ were
parallel with each other. In the present case, the geometries of the orien-
tational isomers were fully optimized without any symmetry restriction.
Interaction energies (DE) that corresponded to the optimized aromatic
dimers (parallel ›› and antiparallel ›fl orientation) were evaluated by
DE=Edimerꢀ2Emonomer at the MP2/6-31G* level, in which Edimer and
Emonomer are energies of the aromatic dimer and the monomer, respective-
ly. DE values were corrected for basis set superposition error (BSSE) by
means of the counterpoise correction of Boys and Bernardi.[38]
[15] a) A. Chauhan, N. Jain, P. Diwan, A. Khopade, J. Drug Targeting
2004, 12, 575–583; b) S. Tenjarla, P. Puranajoti, R. Kasina, T.
Mandal, J. Pharm. Sci. 1998, 87, 425–429.
[16] H. M. Cabral Marques, J. Hadgraft, I. W. Kellaway, Int. J. Pharm.
1990, 63, 259–266.
[17] a) K. S. S. P. Rao, S. M. Hubig, J. N. Moorthy, J. K. Kochi, J. Org.
Chem. 1999, 64, 8098–8104; b) T. Tamaki, T. Kokubu, K. Ichimura,
Tetrahedron 1987, 43, 1485–1494.
[18] G. Grabner, K. Rechthaler, B. Mayer, G. Kohler, K. Rotkiewicz, J.
Phys. Chem. A 2000, 104, 1365–1376.
[19] a) R. A. Agbaria, E. Roberts, I. M. Warner, J. Phys. Chem. 1995, 99,
10056–10060; b) Y. Chen, Y. Tao, H. Lin, Macromolecules 2006, 39,
8559–8566.
Acknowledgements
We gratefully thank the technical support from Annegret Engelke and
Mathias Grosser. We would also like to express our appreciation to
Dagmar Auerbach for her assistance in fluorescence spectroscopy meas-
urements and Yi Dong and Michael Springborg from Saarland University
for helpful discussions concerning molecular modeling. Financial support
from Saarland University and China Scholarship Council is gratefully ac-
knowledged.
[20] N. Miyakawa, T. Kiba, S. I. Sato, Mol. Cryst. Liq. Cryst. 2010, 520,
469–476.
[21] a) K. Kano, H. Matsumoto, S. Hashimoto, M. Sisido, Y. Imanishi, J.
Am. Chem. Soc. 1985, 107, 6117–6118; b) T. Yorozu, M. Hoshino,
M. Imamura, J. Phys. Chem. 1982, 86, 4426–4429.
[1] a) G. Wenz, Angew. Chem. 1994, 106, 851–870; Angew. Chem. Int.
Ed. Engl. 1994, 33, 803–822; b) G. Wenz, J. Polym. Sci. Part A 2009,
47, 6333–6341; c) J. Szejtli, Chem. Rev. 1998, 98, 1743–1753.
[2] a) A. Steffen, C. Thiele, S. Tietze, C. Strassnig, A. Kamper, T. Len-
gauer, G. Wenz, J. Apostolakis, Chem. Eur. J. 2007, 13, 6801–6809;
b) G. Wenz, B. H. Han, A. Mꢀller, Chem. Rev. 2006, 106, 782–817;
c) G. Wenz, C. Strassnig, C. Thiele, A. Engelke, B. Morgenstern, K.
Hegetschweiler, Chem. Eur. J. 2008, 14, 7202–7211.
[3] a) K. Uekama, F. Hirayama, T. Irie, Chem. Rev. 1998, 98, 2045–
2076; b) A. R. Hedges, Chem. Rev. 1998, 98, 2035–2044; c) T. Lofts-
son, M. Brewster, J. Pharm. Sci. 1996, 85, 1017–1025; d) K. Kano,
Y. Ishida, Angew. Chem. 2007, 119, 741–744; Angew. Chem. Int. Ed.
2007, 46, 727–730.
[4] a) I. Beꢄ, M. G. Gotsev, P. M. Ivanov, C. Jaime, P. A. Kollman, J.
Org. Chem. 2006, 71, 2056–2063; b) Y. Liu, B.-H. Han, H.-Y.
Zhang, Curr. Org. Chem. 2004, 8, 35–46; c) K. Kano, Y. Ishida,
Chem. Asian J. 2008, 3, 678–686.
[5] a) S. Filippone, F. Heimann, A. Rassat, Chem. Commun. 2002,
1508–1509; b) T. Loftsson, D. Hreinsdottir, M. Masson, Int. J.
Pharm. 2005, 302, 18–28.
[22] A. Mꢀller, G. Wenz, Chem. Eur. J. 2007, 13, 2218–2223.
[23] K. Harata, Bull. Chem. Soc. Jpn. 1987, 60, 2763–2767.
[24] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, A. Laham, C. Y. Peng, A. Nanayakkara,
2398
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2390 – 2399

Polycyclic Aromatics
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
C. Gonzalez, J. A. Pople, Gaussian , Inc., Wallingford, CT, 2004.
[25] C. Gonzalez, E. C. Lim, J. Phys. Chem. A 2000, 104, 2953–2957.
[26] S. Grimme, Angew. Chem. 2008, 120, 3478–3483; Angew. Chem. Int.
Ed. 2008, 47, 3430–3434.
[27] K. Guckian, B. Schweitzer, X. Rex, C. Sheils, D. Tahmassebi, E.
Kool, J. Am. Chem. Soc. 2000, 122, 2213–2222.
[28] J. Mizoguchi, Y. Kawanami, T. Wada, K. Kodama, K. Anzai, T.
Yanagi, Y. Inoue, Org. Lett. 2006, 8, 6051–6054.
[29] A. W. Snow, E. E. Foos, Synthesis 2003, 0509–0512.
[30] H. Lund, J. Bjerrum, Chem. Ber. 1931, 64, 210–213.
[31] a) S. Nigam, G. Durocher, J. Phys. Chem. 1996, 100, 7135–7142;
b) G. Cox, N. Turro, N. Yang, M. Chen, J. Am. Chem. Soc. 1984,
106, 422–424.
Acta Crystallogr. Sect. C 1984, 40, 428–431; c) V. Petrꢅcek, I. Cꢅsar-
ovꢆ, L. Hummel, J. Kroupa, B. Brezina, Acta Crystallogr. Sect. B
1990, 46, 830–832; d) A. C. Blackburn, L. J. Fitzgerald, R. Gerkin,
Acta Crystallogr. Sect. C 1996, 52, 2862–2864; e) J. Oddershede, S.
Larsen, J. Phys. Chem. A 2004, 108, 1057–1063; f) J. Robertson, H.
Shearer, G. Sim, D. Watson, Acta Crystallogr. 1962, 15, 1–8; g) C.
Brock, J. Dunitz, Acta Crystallogr. Sect. B 1990, 46, 795–806; h) T.
Welberry, Proc. R. Soc. London Ser. A 1973, 334, 19–48.
[35] K. Lindner, W. Saenger, Carbohydr. Res. 1982, 99, 103–115.
[36] a) J. J. P. Stewart, J. Comput. Chem. 1989, 10, 209–220; b) J. J. P.
Stewart, J. Comput. Chem. 1989, 10, 221–264.
[37] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996, 14, 33–
38.
[38] S. F. Boys, F. Bernardi, Mol. Phys. 1970, 19, 553–566.
[32] M. E. Brewster, R. Vandecruys, J. Peeters, P. Neeskens, G. Verreck,
T. Loftsson, Eur. J. Pharm. Sci. 2008, 34, 94–103.
Received: March 1, 2011
[33] H. Takeru, L. L. John, J. Am. Pharm. Assoc. 1954, 43, 349–354.
[34] a) D. Holmes, S. Kumaraswamy, A. Matzger, K. Vollhardt, Chem.
Eur. J. 1999, 5, 3399–3412; b) J. Bouwstra, A. Schouten, J. Kroon,
Published online: July 7, 2011
Chem. Asian J. 2011, 6, 2390 – 2399
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2399