Binding studies on CB[6] with a series of 1-Alkyl-3-methylimidazolium ionic liquids in an aqueous system

Article abstract of DOI:10.1002/asia.200900510

The host-guest chemistry between a series of 1-alkyl-3-methyl-imidazolium bromide ([C_nmim]Br) guests and the macrocyclic host molecule cucurbit[6]uril (CB[6]) in an aqueous system is systematically studied in neutral aqueous media. Both 1D and 2D NMR experiments in conjunction with isothermal titration calorimetry (ITC) unveil the binding characteristics of the host-guest interaction. Solution binding constants (K_a) up to 10⁵m^-1 are measured directly. Additionally, this [C _nmim]Br-CB[6] interaction was found to significantly increase the solubility of CB[6] in neutral water, in some cases by at least four orders of magnitude. From these studies, a detailed host-guest binding model has been constructed and is fully discussed. In this model, the delocalized positive charge on the imidazolium ring becomes partially localized on either one of the nitrogen atoms upon complexation with CB[6]. Localization of the positive charge is directly related to the length of the "1-alkyl" chain on the imidazolium ring, which causes an induced local dipole subsequently allowing for an ion-dipole interaction with the carbonyl portal of CB[6].

Full text of DOI:10.1002/asia.200900510

FULL PAPERS
DOI: 10.1002/asia.200900510
Binding Studies on CB[6] with a Series of 1-Alkyl-3-methylimidazolium Ionic
Liquids in an Aqueous System
Nan Zhao, Li Liu, Frank Biedermann, and Oren A. Scherman*[a]
Dedicated to the 150th anniversary of Japan–UK diplomatic relations
Abstract: The host–guest chemistry be-
tween a series of 1-alkyl-3-methyl-imi-
dazolium bromide ([Cnmim]Br) guests
and the macrocyclic host molecule cu-
curbit[6]uril (CB[6]) in an aqueous
system is systematically studied in neu-
tral aqueous media. Both 1D and 2D
NMR experiments in conjunction with
isothermal titration calorimetry (ITC)
unveil the binding characteristics of the
host–guest interaction. Solution bind-
ing constants (Ka) up to 105 mÀ1 are
measured directly. Additionally, this
[Cnmim]Br–CB[6]
interaction
was
In this model, the delocalized positive
charge on the imidazolium ring be-
comes partially localized on either one
of the nitrogen atoms upon complexa-
tion with CB[6]. Localization of the
positive charge is directly related to
the length of the “1-alkyl” chain on the
imidazolium ring, which causes an in-
duced local dipole subsequently allow-
ing for an ion–dipole interaction with
the carbonyl portal of CB[6].
found to significantly increase the solu-
bility of CB[6] in neutral water, in
some cases by at least four orders of
magnitude. From these studies, a de-
tailed host-guest binding model has
been constructed and is fully discussed.
Keywords: host–guest systems
macrocycles · NMR spectroscopy ·
ionic liquids
chemistry
·
·
supramolecular
Introduction
supramolecular building block.[3–7] While some of the larger
CB[n] macrocyclic hosts[8] such as CB[7], CB[8], CB[10],
and CB[n] derivatives[9] have received increasing interest,
CB[6] is still studied on account of its economical and facile
synthesis and purification.
Although CB[6] has very poor solubility in water
(ꢀ0.018 mm),[10] host-guest binding studies in acidic solution
(HCO2H/H2O (1:1 v/v)) have been successfully conducted
since the 1980s.
Cucurbit[6]uril (CB[6]) is a unique macrocyclic host mole-
cule consisting of six glycoluril units and twelve methylene
bridges (Figure 1).[1,2] Its rigid hydrophobic cavity and two
hydrophilic portals make CB[6] an interesting and promising
Buschmann et al. reported weak binding between CB[6]
and aliphatic alcohols, acids, and nitriles in HCO2H/H2O
(1:1).[11] Binding constants as low as 102–103 mÀ1 were report-
ed and no sign of guest selectivity was observed. On the
other hand, Mock and co-workers reported that charged am-
monium molecules are good guests for CB[6].[12] The posi-
tive charge of ammonium interacts through ion–dipole inter-
actions with the CB[6] portal carbonyl groups, thereby stabi-
lizing the host–guest complexes. For mono-alkyl ammonium
cations, the binding affinity varied from 102 to 105 mÀ1, indi-
cative of guest selectivity for different chain lengths.[13]
Kim et al. reported that CB[6] could also be dissolved in
aqueous saline solutions.[14] X-ray crystallography indicated
that multiple sodium ions interacted with the CB[6] portal
carbonyls acting as a “lid” on the CB[6] barrel to control
Figure 1. The molecular structure of CB[6].
[a] N. Zhao, Dr. L. Liu, F. Biedermann, Dr. O. A. Scherman
Melville Laboratory for Polymer Synthesis
Department of Chemistry
University of Cambridge
Cambridge, CB2 1EW (UK)
Fax : (+44)1223-334866
E-mail: oas23@cam.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.200900510.
530
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 530 – 537

the binding and release of THF molecules from within the
hydrophobic cavity. Buschmann measured the binding con-
stants of CB[6] with a variety of alkali-metal, alkaline-earth,
transition-metal, and lanthanide cations. Their binding con-
stants are in the range of 102–103 mÀ1 in acidic solution.[15–21]
It is important to note that in acidic solution, protons are in
constant competition with the metal cations, therefore, the
actual binding constants between CB[6] and metal cations
are likely to be higher than the measured values report-
ed.[21,22]
Recently, we reported that 1N,3N-dialkylimidazolium, can
act as a new guest and simultaneously serve to dissolve
CB[6] in neutral water without the assistance of metal cat-
ions.[23] 1N,3N-dialkylimidazolium is a room-temperature
ionic liquid (RTIL) which has attracted attention as a new
solvent on account of its outstanding physical-chemical
properties.[24,25] The properties of imidazolium RTILs can be
readily tuned by simply mixing with different organic sol-
vents or water. Many chemical reactions have been carried
out in imidazolium–water mixtures. One example in particu-
lar was reported by Rault-Berthelot et al. where hydrogen
could be generated in situ from the imidazolium–water mix-
ture.[26]
Within the realm of CB chemistry, imidazolium RTILs
are promising guests as their physical properties can be al-
tered in a straightforward manner by varying both counter
ion and the chemical nature of the alkyl substituents. These
alterations yield a large variety of guests which are both
simple to design and prepare.[27] A major difference between
imidazolium cations and alkyl ammonium cations is that the
positive charge on the former can be delocalized over 2.5 ꢁ
between the two nitrogen atoms in the five membered ring
as illustrated in Figure 2.[28]
with CB[7].[33] Sindelar et al. reported two crystal structures
for ethyl and butyl methylimidazolium with CB[6] which
verify our proposed flexible binding model.[34] Very recently,
we reported that 1N,3N-dimethylimidazolium methylsulpho-
nate could be utilized to isolate a lid-free and charge-free
CB[6]–diethyl ether inclusion complex.[35] However, a sys-
tematic study of CB[6] binding with different N1-alkyl chain
lengths has not been probed until now. Herein, we investi-
gate how varying the N1 alkyl chains from ethyl to octyl as
well as dodecyl, affect CB[6] complexation. We also propose
a complete binding model which can serve as a guideline for
future work in this area.
Results and Discussion
A series of imidazolium salts ranging from 1-ethyl-3-methyl-
imidazolium bromide ([C2mim]Br) to 1-octyl-3-methylimida-
zolium bromide ([C8mim]Br) and 1-dodecyl-3-methylimida-
zolium bromide ([C12mim]Br) were synthesized and com-
plexed with CB[6] in aqueous solution. While the counter
ion can certainly have a dramatic effect on solubility and
complexation, these effects are beyond the scope of this
study and will be reported elsewhere. In order to eliminate
any effects from counter-ion binding, all imidazolium guest
molecules in this study contain a bromide counter ion. 1D,
2D NMR spectroscopy, and ITC characterization techniques
were used to study the binding dynamics.
In NMR studies of the CB[6]-imidazolium systems, at
least three factors influence the guestꢂs chemical shifts:
charge localization on the imidazolium ring, guest location
inside the CB[6] cavity, and the rates of complexation and
decomplexation. These factors are interdependent in deter-
mining the final guest chemical shift values. As the effects
of complexation and decomplexation rates are well under-
stood, only the first two effects are discussed below.
The positive charge on an imidazolium ring has an induc-
tive effect on nearby atoms. When the charge partially local-
izes at the N3 atom (Figure 3a), the chemical shifts of Ha
and Hc will move downfield owing to greater withdrawing of
electron density. In addition, Hd and He will shift upfield.
When the positive charge is partially located at the N1 atom
(Figure 3b), the contrary is observed. For Hb, no charge
effect is observed. Thus, the chemical shift variation of Hb
arises from its relative location within CB[6], providing an
insight into the binding geometry of the host–guest complex.
Figure 2. Resonance of an imidazolium cation.[28]
Several other groups have become interested in CB–imda-
zolium host–guest systems as well. Istvan et al. investigated
the binding properties of imidazolium with CB[7].[29] Mac-
artney et al. reported a decrease in H/D exchange rate for
the acidic C2 proton on the imidazolium ring upon complex-
ing with CB[7].[30] Schmitzer et al. found that 1N,3N-disub-
stituted methylenediimidazoli-
um salts complex with CB[7] as
well as several other host mole-
cules.[31] Some of those diimida-
zolium guests were observed to
form ternary complexes with
CB[7]
and
cyclodextrins
(CDs),[32] and some polyimida-
zolium cations containing aro-
matic binding sites could form
[2]- and [3]-pseudorotaxanes Figure 3. a) Positive charge partially located on N3, and b) positive charge partially located on N1.
Chem. Asian J. 2010, 5, 530 – 537
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
531

O. A. Scherman et al.
FULL PAPERS
The cavity of CB[6] provides high electron density which
shields protons inside. The urea groups at the portal regions
also cause magnetic shielding to the proximal guest protons
by the presence of p electrons. Taking into account these ef-
fects, the shielding (deep grey) and deshielding (light grey)
regions of CB[6] are shown in Figure 4.
Therefore, several 2D NMR techniques were employed to
confirm our proton assignments.
Figure 6a and b are different regions of HMBC spectrum
of [C2mim]Br in water. Ha (singlet) and He (quartet) are
easily assigned as shown. Ha has a larger correlation with Cc
Figure 4. Shielding and deshielding regions of CB[6].
Figure 6. (a) and (b) HMBC spectra of [C2mim] Br. (c) and (d) HMBC
spectra of [C2mim]Br in the presence of CB[6].
1D and 2D NMR Studies
1
Figure 5 is a stack plot of H NMR spectra of CB[6] with
[C2mim]Br in neutral water. This host–guest system is in fast
exchange. When binding with CB[6], He and Hf of
than Cd (J3,CH >J4,CH) while He has a larger correlation with
Cd (Figure 6b). As a result, Cc and Cd are assigned and Cc
(124 ppm) is shifted about 2 ppm downfield from Cd
(122 ppm). In Figure 6a, Cc correlates with Hd and Cd corre-
lates with Hc, allowing for the unambiguous assignment of
1
Hd and Hc in the H NMR spectra in Figure 5. When CB[6]
was added into the system (guest to host ratio is 3:1 accord-
ing to integration), the same method was used to assign Hc
and Hd. Different regions of the HMBC spectrum are
shown in Figure 6c and d. Results show that Hd (6.98 ppm)
now moves upfield of Hc (7.30 ppm).
NOESY was also employed to independently verify the
correct assignment of Hc and Hd. Figure 7a is a region of
the NOESY spectrum of [C2mim]Br in water. Compared to
Hd, Hc is much closer to Ha while He is closer to Hd. There-
fore, once Ha and He have been determined, Hd and Hc are
easily assigned. Figure 7b shows the NOESY signals of
[C2mim]Br binding with CB[6]. The same method was em-
ployed to assign Hc and Hd and the result is consistent with
that of the HMBC method discussed above.
Figure 5. 1H NMR stack plot of CB[6] and [C2mim]Br in D2O with the
“intended” ratios of 0, 0.2, 0.4, 0.6, 0.8, and 1.0 equivalent of CB[6].
1
Figure 8 is a stack plot of H NMR spectra of CB[6] with
[C2mim]Br move upfield which indicate that the ethyl chain
on N1 goes inside the CB[6] cavity. The upfield shift of Hb
suggests that at least half of the five member ring is embed-
ded inside CB[6]ꢂs shielding region. Unfortunately, the Hc/
Hd peaks are difficult to distinguish from 1D NMR alone.
[C3mim]Br in water. This host–guest system exhibits fast ex-
change and the N1 alkyl chain was buried inside CB[6]ꢂs
cavity as indicated by protons He and Hf. However, in con-
trast to the CB[6]–[C2mim]Br system, Hb and Hd on the imi-
dazolium ring move downfield while Hc moves upfield. For
532
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 530 – 537

Binding Studies on CB[6] with a Series of Ionic Liquids
Figure 7. NOESY spectra of a) [C2mim]Br and b) [C2mim]Br binding
with CB[6].
Figure 9. 1H NMR stack plots of CB[6] and [C12mim]Br in D2O with the
“intended” ratios of 0, 0.2, 0.4, 0.6, 0.8 and 1.0 equivalent of CB[6].
Values in brackets are the real ratios from integration under intended
ratio.
from free guest was recorded and listed in Table 1. These
numbers indicated the shielding or deshielding effects on
the guest protons upon binding with CB[6]. Negative values
indicate shielding upon binding, while positive values indi-
cate deshielding. In Table 1, [H]r/[G]r shows the real ratio in
solution based on integration of the 1H NMR spectrum
while the intended ratio was 1:1. These ratios indicate the
guestsꢂ ability to solubalize CB[6] into water. With the same
guest concentration, [C4mim]Br and [C5mim]Br can bring an
equimolar amount of CB[6] into solution, showing strong
binding affinities between host and guest. When the N1
chain lengths increase or decrease, the host concentration in
solution decreases and the binding mode is always fast ex-
change, which correlates with a weaker host–guest interac-
tion.
In each case, protons on the N1 alkyl chain move upfield
indicating cavity binding. It is important to mention that for
[C7mim]Br and [C8mim]Br, the upfield shifts of Hf (the ter-
minal CH3 protons) indicate that the whole heptyl or octyl
chain is inside CB[6]. As the heptyl and octyl chains are
longer than the depth of the CB[6] cavity, a conformational
rearrangement of the alkyl chain inside the CB[6] cavity
must take place. For the chain end of [C12mim]Br, a slight
downfield variation was observed for Hf indicating the chain
end likely lies outside the cavity of CB[6].
Figure 8. 1H NMR stack plots of CB[6] and [C3mim]Br in D2O with the
“intended” ratios of 0, 0.2, 0.4, 0.6, 0.8 and 1.0 equivalent of CB[6].
Values in brackets are the real ratios from integration under intended
ratio.
the remainder of the [Cnmim]Br series binding with CB[6]
(for 1H NMR titration plots and typical HMBC and
NOESY spectra of these CB[6]-imidazolium bindings, see
Supporting Information), the imidazolium protons shift in
an analogous manner to those for [C3mim]Br except in the
case of [C12mim]Br.
1
Figure 9 shows the stack plot of H NMR spectra of CB[6]
with [C12mim]Br in water. Besides the fast exchange mode,
it is noticeable that proton Hb shifts upfield as in the case of
[C2mim]Br. Some of the protons on the N1 alkyl chain shift
downfield while other protons shift upfield, clearly indicat-
ing that the alkyl chain is not fully buried inside CB[6]ꢂs
cavity. A conformational rear-
rangement of the alkyl chain is
Table 1. Complexation induced chemical shift data in ppm for [Cnmim]Br–CB[6] in D2O at 258C.
likely to occur, similar to what
has been observed previously
by Kim and co-workers for the
A
Ha
Hb
Hc
Hd
He
Hf
[H]r/[G]r
exchange mode
A
À0.10
0.02
0.07
À0.76
0.04
0.40
0.32
0.14
0.11
0.11
À0.10
À0.16
À0.16
À0.14
À0.06
À0.05
À0.10
À0.54
0.27
0.58
0.35
0.14
0.10
0.09
0.10
À0.46
À0.73
À0.84
À0.39
À0.02
À0.01
À0.05
À0.01
À0.27
À0.50
À0.70
À0.60
À0.48
À0.21
À0.13
0.12
0.46
0.70
1
fast
fast
slow
slow
intermediate
fast
fast
larger CB[8] homologue.[36]
A
R
H
more detailed investigation of
these dynamics is currently un-
derway.
Under a 1:1 intended guest/
host ratio, the chemical shift
N
0.05
1
[a]
N
–
0.91
0.74
0.65
0.51
N
0.02
N
À0.02
[a]
[a]
G
À0.05
–
–
fast
variation of the bound guest [a] Value could not be measured.
Chem. Asian J. 2010, 5, 530 – 537
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
533

O. A. Scherman et al.
FULL PAPERS
From a thermodynamic point of view, the binding process
is largely enthalpy driven. When the alkyl chains of
[Cnmim]Br are longer than n=5, the value of association
enthalpy decreases as illustrated in Table 2. A possible
reason for this observed trend is that CB[6] has an intera-
tomic distance of 6 ꢁ between carbonyl oxygens axially
spanning the cavity, which can optimally be filled by 4–5
methylene groups.[37] Consequently, [C4mim]Br and
[C5mim]Br fit best inside the CB[6] cavity. For shorter
chains, the hydrophobic effect is weaker, while for longer
chains, the chain end, if fully extended, will exit the CB[6]
cavity or require a backfolded conformation,[36] both of
which are energetically unfavorable.
From the ITC experiments, the binding stoichiometry of
[C4mim]Br, [C5mim]Br, and [C6mim]Br with CB[6] were all
1:1. In the absence of any buffers or salts,[34] [C2mim]Br
forms a 2:1 complex in solution with CB[6] as we reported
previously.[23] Whether [C3mim]Br will adopt a 2:1 binding
stoichiometry with CB[6] is still unknown.
Host–Guest Binding Constants and Stoichiometry Studies
In host–guest chemistry, both binding stoichiometry and as-
sociation constants are crucial to the understanding of host–
guest binding events. The most common methods to deter-
mine these are titration experiments using spectroscopic
1
methods such as H NMR, UV/Vis and fluorescence.[38] As
the CB[6]-imidazolium UV/Vis absorption mechanism is not
1
yet understood, H NMR titration methods were employed
as a first attempt to understand the binding. Unfortunately,
repeated titrations in an attempt to construct a Job plot
proved impossible. This was likely to arise from the insolu-
bility of CB[6] in the absence of the imidazolium guest, pre-
venting us from obtaining any starting points. Fortunately,
ITC was found to be a useful and relatively convenient
method to obtain both the binding stoichiometry and associ-
ation constants.
In a typical ITC experiment, CB[6] and [Cnmim]Br were
dissolved in separate 10 mm phosphate buffer solutions
(pH 7) and the binding experiments were carried out at
258C. Binding stoichiometry and thermodynamic data were
obtained and are listed in Table 2 depicting a clear trend of
The binding constants and thermodynamic data of CB[6]
and monoalkyl ammonium species with different chain
lengths were measured by Buschmann (Table 2).[37] Al-
though Buschmannꢂs experi-
ments were done in 50%
Table 2. Binding constants and thermodynamic data for [Cnmim]Br–CB[6] in phosphorate buffer solution at
258C and mono-ammonium-CB[6] in 50% formic acid at 258C.[37]
formic acid, it has been suggest-
ed that the binding constants
are nearly independent from
the solvent acidity. The same
trends were observed for both
binding constants and enthalpy
changes in our study with alkyl
imidazoliums. While the highest
binding constant for CB[6] with
A
DG [kJmolÀ1
]
DH [kJmolÀ1
]
TꢃDS [kJmolÀ1
]
log Ka [LmolÀ1
]
stoichiometry
2:1[23]
[a]
[a]
[a]
E
–
–
–
–
–
<4
<4
5.1
5.4
4.5
<4
<4
<4
2.73
3.54
4.05
3.81
3.83
2.67
[a]
[a]
[a]
[a]
N
–
–
N
À29.3
À30.5
À25.9
À30.0
À31.1
À19.1
À0.7
À0.6
6.9
1:1
1:1
1:1
N
G
[a]
[a]
[a]
[a]
N
–
–
–
–
–
–
–
–
–
–
[a]
[a]
[a]
[a]
G
–
[a]
[a]
[a]
[a]
A
–
+[37]
the
series was found to be 104 mÀ1
for CH3A
(CH2)3NH3+, the high-
monoalkyl
ammonium
CH3CH2NH3
À15.6
À21.2
À29.3
À32.0
À23.0
À15.2
À4.6
11.0
7.0
À2.5
À4.6
0.9
1:1
1:1
1:1
1:1
1:1
1:1
+[37]
CH
CH
CH
CH
CH
3A
3A(CH2)3NH3
3A(CH2)4NH3
3A(CH2)5NH3
3A(CH2)6NH3
À14.2
À26.8
À27.4
À22.1
À9.6
+[37]
+[37]
+[37]
+[37]
E
CTHUNGTRENNUNG
CHTUNGTRENNUNG
est binding constant in our
study is considerably higher,
105.35 MÀ1 for [C5mim]Br. This
suggests that for the same alkyl
chain length, the imidazolium
CHTUNGTRENNUNG
G
5.6
[a] Value could not be measured
host–guest binding strengths. For [C2mim]Br and
[C3mim]Br, the binding constants are relatively weak and es-
timated to be lower than 104 mÀ1 which is consistent with fast
exchange observed by 1H NMR. For [C4mim]Br and
[C5mim]Br, the binding strengths are strong (above 105 mÀ1)
series have a higher binding affinity for CB[6].
Developing a Binding Model
1
Systematic movement of the chemical shifts in the H NMR,
especially for protons Hc and Hd, verified two binding
models (Figure 10). For [C2mim]Br, the positive charge is
partially located on the N3 atom, as shown in Figure 10a.
The strong upfield shift of proton Hb indicates that a large
part of the guest molecule (Hd, He, and Hf) has been embed-
ded inside CB[6]ꢂs shielding region.
1
and are in slow exchange in H NMR as discussed above.
The binding constant for [C6mim]Br is 104.5 mÀ1 correspond-
ing to an intermediate exchange on the NMR time scale. By
extending the alkyl chain length to [C7mim]Br, the binding
strength is again too weak to be measured by ITC and ap-
1
pears to return to fast exchange in the H NMR. While high
binding constants do not always indicate slow exchange in
NMR, such as in the case of Istvanꢂs work,[29] in our CB[6]-
imidazolium system, the exchange rates are consistent with
binding affinities.
Both nearest neighbor protons to N1, He on the alkyl
chain and Hd on the imidazolium ring, were (on average)
shifted upfield; this was caused by both the cavity shielding
effect of CB[6] and the positive charge shielding effect from
the localized charge existing on N3. Conversely, Ha on the
534
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 530 – 537

Binding Studies on CB[6] with a Series of Ionic Liquids
1D and 2D NMR techniques as well as ITC. A clear trend
in binding with the length of the alkyl group on the N1 imi-
dazolium position was observed. Upon binding, the N1 alkyl
chain enters the hydrophobic cavity of CB[6] to maximize
van der Waals interactions and the hydrophobic effect, while
the positive charge on the imidazolium ring remains at the
CB[6] portal region to maximize the ion–dipole interactions.
For [C2mim]Br, the positive charge of the imidazolium ring
is partially located on N3 to maximize both of these effects,
however, as the ethyl chain is relatively short, a binding con-
stant <104 mÀ1 and fast exchange on the NMR time scale
were observed. From [C3mim]Br to [C8mim]Br, the positive
charge on the ring becomes partially located on N1 as indi-
cated by 2D NMR experiments. It was found that the tight-
est fit existed for butyl and pentyl alkyl chains; consequent-
ly, [C4mim]Br and [C5mim]Br have the strongest binding af-
finities with CB[6] (Ka >105 mÀ1) and exhibit slow exchange
on the NMR time scale. While longer alkyl chains still bind
inside the CB[6] cavity, the energetics for this complexation
are somewhat unfavorable and result in decreased binding
affinities and intermediate and/or fast exchange on the
NMR time scale. Finally, for substantially longer N3 alkyl
chains, as in [C12mim]Br, the NMR behavior of its imidazoli-
um ring resembles that of [C2mim]Br and a conformational
rearrangement of the alkyl chain is likely to occur. This sys-
tematic investigation of alkyl chain penetration depth,
charge localization, and a binding model for imidazolium-
based guests inside CB[6] should serve as a useful guide for
both synthetic and supramolecular chemists in designing and
preparing novel functional self-organizing systems with
CB[6] in neutral water.
Figure 10. Two binding models of CB[6] with imidazolium.
N3 methyl group and Hc of the imidazolium ring, should ex-
perience a downfield shift caused by the positive charge in-
ductive effect, however, these are not observed in the
1H NMR spectrum. In fact, Hc shifts slightly upfield and Ha
remains almost unchanged. The lack of large chemical shifts
in these cases can be rationalized by the downfield shifts
being balanced out by shielding effects owing to the pene-
tration depth for the smaller [C2mim]Br within the cavity of
CB[6]. When considering a larger dialkylimidazolium guest,
[C3mim]Br, the positive charge localizes closer to N1 in
order to optimize both the hydrophobic effect and ion–
dipole interactions as indicated in Figure 10b. Proton Hb ex-
hibits a downfield shift suggesting that at least half of the
imidazolium ring (N3 and C4 with Hc) remains outside the
portal region of CB[6] and thus is in the deshielding region.
Compared with [C2mim]Br, the larger [C3mim]Br does not
penetrate the CB[6] host as deeply. In this host–guest
system, the positive charge should cause Hd and He to shift
downfield while inducing an upfield shift in Ha and Hc. As
the alkyl chain remains inside the CB[6] cavity, a strong
shielding effect is dominant and He (and all other alkyl pro-
tons) experience an upfield shift. However, as Hd is located
around the portal region, the shielding effect is not as strong
as inside the cavity and the overall chemical shift is domi-
nated by charge localization and the peak moves downfield.
For Ha and Hc, the deshielding effect of CB[6] seems to
dominate again over the positive charge effect as they both
shift downfield. For [C4mim]Br through [C8mim]Br, the up-
field and downfield shifts from Ha to Hf are identical to
those observed for [C3mim]Br as described above.
Experimental Section
General
All chemicals used were of reagent grade or higher and used without fur-
ther purification unless otherwise stated. 1-bromopentane (98%) and 1-
bromohexane (98%) were obtained from Acros Organics. Ammonium
hexafluorophosphate and tetraoctylammonium bromide were purchased
from Alfa Aesar. 1-Butyl-3-methylimdazolium (97%) was supplied by
Fluka. Tetrahydrofuran, dichloromethane, and diethyl ether were ob-
tained from Fisher Scientific. 1-Methylimidazole (99%) was purchased
from Lancaster. 1-Bromopropane (98%) was supplied by Avocado Or-
ganics. 1-Bromoheptane and 1-bromo-dodecane (96%) were purchased
from Aldrich Chemicals. CB[6] was prepared using a one-pot reaction
method.[8]
When the length of the alkyl chain becomes much longer,
as in [C12mim]Br, the proton Hf of the dodecyl chain end ap-
peared to no longer reside inside the cavity of CB[6] evi-
denced by a slight downfield shift. On the other hand,
proton Hb on the imidazolium ring moves upfield indicating
the five-membered ring resides deeply inside the shielding
area of CB[6], similar to [C2mim]Br.
Synthesis
General Procedure for [Cnmim]Br Synthesis: For [C2mim]Br: 1-Methyli-
midazole (1.6 g, 20 mmol) and 1-bromoethane (4.5 g, 25 mmol) were re-
fluxed in THF(tetrahydrofuran) for 72 h. The solvent was decanted and
the residue was washed with diethyl ether and dried under vacuum yield-
1
ing the title compound as a white crystalline solid (3.1 g, 80%). H NMR
(400 MHz, D2O): dH =1.49 (3H, t, J=6 Hz), 3.88 (3H, s), 4.22 (2H, q,
J=6 Hz), 7.42 (1H, t, J=1.6 Hz), 7.49 (1H, t, J=1.6 Hz), 8.71 ppm (1H,
s). Remaining characterization data is consistent with results previously
published.[39]
Conclusions
AHCTUNGTRENNUNG
[C3mim]Br was obtained as a transparent liquid (3.5 g, 86%). 1H NMR
(400 MHz, D2O): dH =0.86 (3H, t, J=6 Hz), 1.83 (2H, q, J=6 Hz), 3.84
(3H, s), 4.11 (2H, t, J=5.6 Hz), 7.38 (1H, t, J=0.6 Hz), 7.42 (1H, t, J=
Binding properties between [Cnmim]Br guest molecules
with the macrocyclic host CB[6] were investigated with both
Chem. Asian J. 2010, 5, 530 – 537
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
535

O. A. Scherman et al.
FULL PAPERS
0.6 Hz), 8.66 ppm (1H, s). Remaining characterization data is consistent
with results previously published.[39]
Acknowledgements
ACHTUNGTRENNUNG
[C5mim]Br was obtained as a yellowish liquid (3.8 g, 80.7%). 1H NMR
N.Z. gratefully acknowledges the Raymond and Helen Kwok Research
Scholarship for financial support. F.B. thanks the German Academic Ex-
change Service (DAAD) for financial support. We thank all members in
the Scherman Group for their helpful discussions.
(400 MHz, D2O): dH =0.80 (3H, t, J=6 Hz), 1.22 (4H, m), 1.80 (2H, m,
J=5.6 Hz) 3.82 (3H, s), 4.12 (2H, t, J=6 Hz), 7.36 (1H, t, J=1.6 Hz),
7.41 (1H, t, J=1.6 Hz), 8.64 ppm (1H, s). Remaining characterization
data is consistent with results previously published.[39]
ACHTUNGTRENNUNG
[C6mim]Br was obtained as a transparent liquid (4.0 g, 80.1%). 1H NMR
(400 MHz, D2O): dH =0.81 (3H, t, J=4.4 Hz), 1.26 (6H, m), 1.85 (2H,
m), 3.88 (3H, s), 4.18 (2H, t, J=5.6 Hz), 7.42 (1H, t, J=1.2 Hz), 7.47
(1H, t, J=1.6 Hz), 8.71 ppm (1H, s). Remaining characterization data is
consistent with results previously published.[40]
[1] R. Behrend, E. Meyer, F. Rusche, Liebigs Ann. Chem. 1905, 339, 1–
37.
[2] W. A. Freeman, W. L. Mock, N. Shih, J. Am. Chem. Soc. 1981, 103,
7367–7368.
[3] N. J. Wheate, Aust. J. Chem. 2006, 59, 354–354.
[4] Y. H. Ko, E. Kim, I. Hwang, K. Kim, Chem. Commun. 2007, 1305–
1315.
[5] J. Lagona, P. Mukhopadhyay, S. Chakrabarti, L. Isaacs, Angew.
Chem. 2005, 117, 4922–4949; Angew. Chem. Int. Ed. 2005, 44, 4844–
4870.
ACHTUNGTRENNUNG
[C7mim]Br was obtained as a transparent liquid (4.2 g, 79.6%). 1H NMR
(400 MHz, D2O): dH =0.809 (3H, t, J=5.2 Hz), 1.23 (8 H ,m), 1.83
(2H,m, J=5.6 Hz), 3.86 (3H, s), 4.15 (2H, t, J=5.6 Hz), 7.40 (1H, d, J=
1.6 Hz), 7.44 (1H, d, J=1.6 Hz), 8.71 ppm (1H, s). Remaining characteri-
zation data is consistent with results previously published.[40]
ACHTUNGTRENNUNG
[C8mim]Br was obtained as a transparent liquid (4.2 g, 79.6%). 1H NMR
[6] K. Kim, N. Selvapalam, D. H. Oh, J. Inclusion Phenom. Macrocyclic
Chem. 2004, 50, 31–36.
(400 MHz, D2O): dH =0.80 (3H, t, J=5.6 Hz), 1.23 (10 H ,m), 1.82
(2H,m, J=5.6 Hz), 3.85 (3H, s), 4.15 (2H, t, J=5.7 Hz), 7.40 (1H, t),
7.43 (1H, t), 8.68 ppm (1H, s). Remaining characterization data is consis-
tent with results previously published.[40]
[7] L. Isaacs, Chem. Commun. 2009, 619–629.
[8] J. Kim, I. Jung, S. Kim, E. Lee, J. Kang, S. Sakamoto, K. Yamaguchi,
K. Kim, J. Am. Chem. Soc. 2000, 122, 540–541.
[9] S. Y. Jon, N. Selvapalam, D. H. Oh, J. Kang, S. Kim, Y. J. Jeon, J. W.
Lee, K. Kim, J. Am. Chem. Soc. 2003, 125, 10186–10187.
[10] H. Buschmann, E. Cleve, K. Jansen, A. Wego, E. Schollmeyer,
Mater. Sci. Eng. C 2001, 14, 35–39.
ACHTUNGTRENNUNG
[C12mim]Br was obtained as a white solid (5.4 g, 80.7%). 1H NMR
(400 MHz, D2O): dH =0.78 (3H, t, J=5.7 Hz), 1.19 (14H, s), 1.27 (4H, s),
1.83 (2H, m), 3.88 (3H, s), 4.20 (2H, t, J=6.2 Hz), 7.48 (1H, d), 7.49
(1H, d), 8.88 ppm (1H, s). Remaining characterization data is consistent
with results previously published.[40]
[11] H. Buschmann, K. Jansen, E. Schollmeyer, Thermochim. Acta 2000,
346, 33–36.
NMR Titration Experiments
[12] W. L. Mock, N. Shih, J. Org. Chem. 1983, 48, 3618–3619.
[13] W. L. Mock, N. Shih, J. Org. Chem. 1986, 51, 4440–4446.
[14] Y. Jeon, J. Kim, D. Whang, K. Kim, J. Am. Chem. Soc. 1996, 118,
9790–9791.
[15] R. Hoffmann, W. Knoche, C. Fenn, H. Buschmann, J. Chem. Soc.
Faraday Trans. 1994, 90, 1507–1511.
[16] H. J. Buschmann, E. Cleve, E. Schollmeyer, Inorg. Chim. Acta 1992,
193, 93–97.
[17] G. Zhang, Z. Xu, S. Xue, Q. Zhu, Z. Tao, Wuji Huaxue Xuebao
2003, 19, 655–659.
[18] H. Buschmann, E. Cleve, K. Jansen, E. Schollmeyer, Anal. Chim.
Acta 2001, 437, 157–163.
[19] H.-J. Buschmann, E. Cleve, K. Jansen, A. Wego, E. Schollmeyer, J.
Inclusion Phenom. Macrocyclic Chem. 2001, 40, 117–120.
[20] X. X. Zhang, K. E. Krakowiak, G. Xue, J. S. Bradshaw, R. M. Izatt,
Ind. Eng. Chem. Res. 2000, 39, 3516–3520.
[21] H. Buschmann, K. Jansen, C. Meschke, E. Schollmeyer, J. Solution
Chem. 1998, 27, 135–140.
[22] R. M. Izatt, R. E. Terry, B. L. Haymore, L. D. Hansen, N. K. Dalley,
A. G. Avondet, J. J. Christensen, J. Am. Chem. Soc. 1976, 98, 7620–
7626.
[23] L. Liu, N. Zhao, O. A. Scherman, Chem. Commun. 2008, 1070–
1072.
[24] T. Welton, Chem. Rev. 1999, 99, 2071–2083.
[25] A. Paul, P. K. Mandal, A. Samanta, J. Phys. Chem. B 2005, 109,
9148–9153.
[26] R. F. de Souza, J. C. Padilha, R. S. Goncalves, J. Rault-Berthelot,
Electrochem. Commun. 2006, 8, 211–216.
[27] P. Nockemann, K. Binnemans, K. Driesen, Chem. Phys. Lett. 2005,
415, 131–136.
[28] P. A. Hunt, B. Kirchner, T. Welton, Chem. Eur. J. 2006, 12, 6762–
6775.
[29] M. Zsombor, B. Laszlo, M. Monika, J. Istvan, J. Phys. Chem. B 2009,
113, 1645–1651.
[30] R. Wang, L. Yuan, D. Macartney, Chem. Commun. 2006, 2908–
2910.
[31] N. Noujeim, L. Leclercq, A. R. Schmitzer, J. Org. Chem. 2008, 73,
3784–3790.
1H NMR spectra were recorded on a Bruker DRX-400. HMBC spectra
were recorded on a Bruker DRX-400 and NOESY spectra were recorded
on a Bruker Avance 500. Chemical shifts are quoted in parts per million
in aqueous solution. Samples with different ratios of CB[6] and
[Cnmim]Br ([H]/[G]) were prepared for titration experiments to reveal
the binding properties between CB[6] and [Cnmim]Br. CB[6] concentra-
tion was kept constant while the concentration of guest was varied with
[H]/[G] ratios of 0.2, 0.4, 0.6, 0.8, and 1.0. The samples were prepared as
follows: 10 mgmLÀ1 [Cnmim]Br solutions were prepared by dissolving
[Cnmim]Br in deuterium oxide. CB[6] (5 mg, 5 mmol) was weighed into a
small sample vial. The calculated amount of imidazolium solution was in-
jected into the same vial. Then deuterium oxide was added to the mix-
ture to keep a total volume of 0.6 mL in all cases. After stirring for one
hour, the sample solutions were transferred into NMR tubes. For 2D ex-
periments, higher concentrations of CB[6] (25 mg, 25 mmol) and corre-
sponding concentrations of the imidazolilum guest were used.
ITC Titration Experiments
Titration experiments were carried out on a VP-ITC from Microcal Inc.
at 25oC in 10 mm sodium phosphate buffer (pH 7).The pH was checked
periodically. Fresh analyte solutions were prepared every few days and
were dissolved by sonication and heating up to 60oC. All solutions were
degassed prior to titration. The binding equilibria of all guests was stud-
ied using a cellular CB[6] concentration of 0.05 mm, to which a 0.6 mm
guest solution was titrated. The guest in the injection syringe was added
at a concentration of 1.0–2.0 mm. Typically 20–30 consecutive injections
of 10–15 mL were added, whereby the first injection was chosen to be
2 mL in all cases. Thus the first data point was removed from the data set
prior to curve fitting. Heats of dilution were found to be negligible in all
cases. For binding constants larger than 105 mÀ1, heats of dilution were de-
termined by titrating beyond saturation and subtracted from the data set.
The data was analyzed with Origin 7.0 software, using the one set of sites
model. A mean value from at least 3 measurements was used to deter-
mine the binding constants and heats of formation, unless otherwise
stated. Reproducibility was tested using different batches of stock solu-
tions of host, guest, and buffer. Deviations were found to be smaller than
6% for all binding constants.
[32] L. Leclercq, N. Noujeim, S. H. Sanon, A. R. Schmitzer, J. Phys.
Chem. B 2008, 112, 14176–14184.
536
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 530 – 537

Binding Studies on CB[6] with a Series of Ionic Liquids
[33] S. Samsam, L. Leclercq, A. R. Schmitzer, J. Phys. Chem. B 2009,
113, 9493–9498.
[37] C. Meschke, H. J. Buschmann, E. Schollmeyer, Thermochim. Acta
1997, 297, 43–48.
[38] C. A. Schalley, Analytical Methods in Supramolecular Chemistry,
Wiley-VCH, 2007.
[39] E. R. Parnham, R. E. Morris, Chem. Mater. 2006, 18, 4882–4887.
[40] J. M. Obliosca, S. D. Arco, M. H. Huang, J. Fluoresc. 2007, 17, 613–
618.
[34] V. Kolman, R. Marek, Z. Strelcova, P. Kulhanek, M. Necas, J. Svec,
V. Sindelar, Chem. Eur. J. 2009, 15, 6926–6931.
[35] L. Liu, N. Nouvel, O. A. Scherman, Chem. Commun. 2009, 3243–
3245.
[36] Y. H. Ko, H. Kim, Y. Kim, K. Kim, Angew. Chem. 2008, 120, 4174–
4177; Angew. Chem. Int. Ed. 2008, 47, 4106–4109.
Received: September 29, 2009
Published online: February 8, 2010
Chem. Asian J. 2010, 5, 530 – 537
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
537