Modes of Micromolar Host-Guest Binding of β-Cyclodextrin Complexes Revealed by NMR Spectroscopy in Salt Water

Article abstract of DOI:10.1021/acs.joc.0c02917

Multitopic supramolecular guests with finely tuned affinities toward widely explored cucurbit[n]urils (CBs) and cyclodextrins (CDs) have been recently designed and tested as functional components of advanced supramolecular systems. We employed various spa

Full text of DOI:10.1021/acs.joc.0c02917

pubs.acs.org/joc
Article
Modes of Micromolar Host−Guest Binding of β‑Cyclodextrin
Complexes Revealed by NMR Spectroscopy in Salt Water
̌
̌
́
Josef Tomecek, Andrea Cablová, Aneta Hromádková, Jan Novotny,* Radek Marek, Ivo Durník,
̌
Petr Kulhánek,* Zdenka Prucková, Michal Rouchal, Lenka Dastychová, and Robert Vícha*
Cite This: J. Org. Chem. 2021, 86, 4483−4496
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Multitopic supramolecular guests with finely tuned affinities toward widely explored cucurbit[n]urils (CBs) and
cyclodextrins (CDs) have been recently designed and tested as functional components of advanced supramolecular systems. We
employed various spacers between the adamantane cage and a cationic moiety as a tool for tuning the binding strength toward CB7
to prepare a set of model guests with KCB7 and Kβ‑CD values of (0.6−5.0) × 1010
M
−1 and (0.6−2.6) × 106 M−1, respectively. These
accessible adamantylphenyl-based binding motifs open a way toward supramolecular components with an outstanding affinity
1
toward β-cyclodextrin. H NMR experiments performed in 30% CaCl2/D2O at 273 K along with molecular dynamics simulations
allowed us to identify two arrangements of the guest@β-CD complexes. The approach, joining experimental and theoretical
methods, provided a better understanding of the structure of cyclodextrin complexes and related molecular recognition, which is
highly important for the rational design of drug delivery systems, molecular sensors and switches.
forms the tightest 1:1 host−guest complex with CB7 (K = 7.2
× 1017 M−1 in D2O) that has ever been reported.5a
INTRODUCTION
■
Cyclodextrins (CDs) and cucurbit[n]urils (CBs) are two
families of supramolecular macrocyclic hosts which can bind
guest molecules into their cavities. Because of the comparable
size of their interior cavities, corresponding homologues bind
guests of similar size and shape.1 However, the binding
strength can differ significantly. Whereas the hydrophobic
effect and hydrogen bonds are the main binding forces in both
families, they can be accompanied by ion−dipole interactions
in the case of CBs.2 The highest affinity toward CDs was
observed for derivatives of cage diamondoids, which perfectly
fit their cavities. For example, the value of association constant
for 1:1 complex of triamantane-9-carboxylic acid and β-CD
reaches the order of 106 M−1 (carbonate buffer, pH = 10.5, 298
K).3 The presence of a positive charge in the guest molecule
usually plays a marginal role in the binding by unmodified
CDs. In contrast, if lipophilic cages like adamantane,4
diamantane,5 bicyclo[2.2.2]octane,4a cubane,6 or ferrocene7
are decorated with cationic moieties, outstanding binding
strengths with geometrically complementary CBs can be
achieved. For example, 4,9-bis(trimethylammonio)diamantane
Analyzing binding data of an extensive series of complexes,
as Nau did in his review in 2015,2 the individual contributions
of the hydrophobic effect and ion−dipole interactions can be
quantified as follows: whereas the hydrophobic effect with the
release of high-energy water molecules from the CB cavity and
desolvation of the guest surface can account for an association
constant as high as 1010 to 1012 M−1, the strength of ion−
dipole interactions is much lower and adds a factor of up to 103
for one cation−portal interaction.
Another feature of inclusion complexes between CB7 and
adamantane derivatives is their tight fit. As a result, the
adamantane cage cannot move significantly from the optimal
Received: December 10, 2020
Published: March 2, 2021
https://dx.doi.org/10.1021/acs.joc.0c02917
© 2021 American Chemical Society
J. Org. Chem. 2021, 86, 4483−4496
4483

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
position inside the cavity8 contrary to derivatives of linear
aliphatic hydrocarbons9 as has been indicated by molecular
dynamic simulations.10 Because adamantane-based ligands are
tightly anchored within the complex and ion−dipole
interactions between the cationic part of the guest and the
CB7 portal cannot surpass the hydrophobic forces,8 the
contribution of the ion−dipole interaction should be easily
adjusted by changing the length of a rigid linker between
adamantane cage and the cationic part. Values of association
constants can then, theoretically, range from 108 to 1015 M−1.
These values were reported for adamantane derivatives with a
long noncharged chain in the position one10 or even for
anionic complexes of Ru(III)8 and singly positively charged
adamantane derivatives with a cationic part in the optimal
position,4a respectively.
a
Scheme 1. Synthesis of Guests 5a and 5b
The design of multitopic guests is an intriguing challenge as
resulting supramolecular assemblies can be used for catalysis,
drug delivery and release,11 or sensor preparation.12 For
example, Yan and co-workers have recently prepared a
rotaxane molecular reactor consisting of a biphenyl axle,
modified γ-CD, and two CB6 stoppers.13 This system catalyzes
photoisomerization of (Z,Z)-1,3-cyclooctadiene to (Z,E)-1,3-
cyclooctadiene with an enantiomeric excess of up to 15.3%. As
an extension of a pioneering work in the 1980s,14 a number of
supramolecular systems based on multitopic guest have been
recently reported to demonstrate importance of lateral
interactions between adjacent macrocycles15 or switching the
arrangements upon chemical signals.16
As a part of our ongoing research on a multitopic guests, we
developed binding motifs with precisely tuned binding
strengths toward CBs and CDs. In this study, we report a
series of adamantane-based guests with a cationic moiety
linked via the benzene ring. We expected a medium binding
strength (i.e., KCB7 ≈ 1010 M−1) toward cucurbit[7]uril (CB7)
and essentially unaffected binding into β-CD (i.e., Kβ‑CD ≈ 105
M−1). Surprisingly, the observed affinities of our new ligands
toward β-CD were of 106 M−1 attacking the strongest 1:1
complexes of β-CD reported so far.3 In addition, we found that
there are two distinct binding modes in the aqueous solution.
As cyclodextrin complexes usually display binding in a fast
exchange regime, obtaining geometry of the complexes is
troublesome. In this account, we introduce a new method-
ology, joining low-temperature NMR and molecular modeling,
which overcomes these limitations and allows for the better-
targeted design of supramolecular systems.
a
Isolated yields are given in parentheses. * and ** indicate that the
reaction was started from a 2a/2b mixture of 19:1 and 1:2,
respectively.
separation. Similarly, the bromo derivatives 3a and 3b were not
separated. However, a single crystal of 3b was obtained from
an oily mixture of both isomers to allow unambiguous
structure determination using NMR and X-ray diffraction
(Figure S41, Table S1). Subsequently, a mixture of 3a and 3b
was treated with sodium imidazolide in DMF, and the
obtained imidazoles 4a and 4b were separated using repeated
column chromatography. It is worth noting that the isolated
yields of pure imidazoles 4 can be improved using a more
efficient separation system or by further purification of mixed
portions since the conversion of 3 to 4 was quantitative,
according to GC−MS and TLC. Finally, imidazoles 4a and 4b
were methylated by MeI to yield imidazolium salts 5a and 5b.
The structure of 5b was verified by a single-crystal X-ray
diffraction analysis (Figure S42, Table S1).
An additional three guests were synthesized via intermediate
4-(1-adamantyl)aniline (8), which was readily prepared from 1
via Friedel−Crafts alkylation of benzene, nitration, and
reduction, as can be seen in Scheme 2. The third
a
Scheme 2. Synthesis of Guests 10−12
RESULTS AND DISCUSSION
■
Chemistry. The nature of the cationic moiety has an
important impact on supramolecular properties of the guests.
Tetraalkylammonium, alkylpyridinium, and dialkylimidazolium
salts are the most popular cationic species extensively studied
in supramolecular chemistry. Therefore, we prepared five
model guests to represent all these classes of compounds. We
started from commercially available 1-bromoadamantane (1),
which was reacted with toluene in the presence of Lewis acid
(Scheme 1). It was previously demonstrated that the meta-
derivative is formed from the para-derivative via subsequent
rearrangement catalyzed by Lewis acids.17 Therefore, we
employed two types of Lewis acids, InCl3 and AlCl3, to
achieve a significantly different ratio of para and meta
regioisomers of 1-adamantyltoluenes 2a and 2b. Due to the
similar chemical properties of the regioisomers, the mixture of
2a and 2b was used in the next bromination step without
a
Isolated yields are given in parentheses. DNPP = N-(2,4-
dinitrophenyl)pyridinium chloride.
4484
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Table 1. Thermodynamic Parameters Obtained by ITC in H2O at 303 K
guest
host
n
K (dm3·mol−1
)
ΔH (kJ·mol−1
−81.7 1.6
)
−TΔS (kJ·mol−1
)
ΔG (kJ·mol−1
−57
−37.23 0.10
−60
−32.54 0.10
−61
−34.42 0.04
−61
−34.22 0.07
−62
−33.71 0.02
)
a
5a
CB7
β-CD
CB7
β-CD
CB7
β-CD
CB7
β-CD
CB7
β-CD
1.00 0.08
1.07 0.04
1.02 0.05
1.01 0.03
0.84 0.05
0.98 0.01
0.97 0.11
0.98 0.03
1.05 0.05
1.06 0.01
(6.1 1.4)×109
24.9 1.5
2.76 0.01
2
(2.61 0.11)×106
−39.99 0.04
−88
a
5b
10
11
12
(2.3 0.6)×1010
2
28
2
2
(4.05 0.17)×105
−36.1 0.8
−83.0 1.5
−36.5 0.4
−80.8 1.3
−35.8 1.7
−76.4 0.8
−33.3 0.5
3.6 0.7
22.1 1.4
2.1 0.4
19.7 1.2
1.6 1.7
14.1 1.0
−0.4 0.5
a
(3.1 0.7)×1010
3
(8.53 0.13)×105
a
(3.4 0.8)×1010
(7.9 0.2)×105
(4.7 1.9)×1010
(6.4 0.7)×105
3
a
3
a
1-Hexyl-3-methylimidazolium chloride was used as a competitor.
imidazolium-derived guest 10, in which the imidazole core is
directly linked to the benzene ring, was prepared according to
a one-pot, four-component methodology and subsequent
methylation by MeI (Scheme 2). The imidazole formation
step was not optimized. Trimethylammonium salt 11 was
prepared via a conventional methylation approach under mild
basic conditions. Finally, pyridinium salt 12 was obtained via
the Zincke reaction.18 Preparation of an analogous triflate salt
of 12 was recently reported.19 However, the para derivative
was accompanied by a significant portion of ortho (33%) and
meta (47%) isomers, and the authors did not separate them.
Before further use, all salts were dried in a vacuum (<1 Torr,
30 °C) and stored under argon atmosphere.
Generally, binding strengths of CDs are significantly lower
than those of CBs due to higher flexibility and a lack of specific
(e.g., ion−dipole) interactions. Complexes of CDs with K
values in order of 105 M−1 are usually considered as very stable.
To our best knowledge, the strongest complex of β-CD was
recently reported by Schibilla and co-workers, who described
the binding behavior of triamantane-9-carboxylic acid.3 This
compound and β-CD subsequently form 1:1 and 1:2
complexes with K1 = 2.95 × 106 M−1 and K2 = 1.4 × 104 M−1.
Interestingly, all of our new guests displayed unexpectedly
high affinities toward β-CD within the range of (0.4−2.6) ×
106 M−1. In comparison with CB7, the trend of K value
dependency on the linker length is opposite in the case of β-
CD complexes (Figure S39). The guest 5a with the longest
spacer provided the highest affinity toward β-CD and vice
versa. The guest 5b stands apart from this trend most likely
due to its bent structure, which does not allow for deeper guest
burying into the β-CD cavity. Interestingly, NMR experiments
revealed two distinct 1:1 binding modes for all examined
guests with β-CD. Therefore, the K values reported in Table 1
should be understood as apparent association constants
averaging the K values of all complexes, which are populated
in equilibria.
As seen in Table 1, formation of all examined complexes was
driven by enthalpic gain, which was accompanied by an
entropic penalty. Figure S40 shows an enthalpy−entropy
scatter plot for our complexes in the context of previously
described high-affinity complexes of CB74a and a compre-
hensive set of cyclodextrin complexes reported by Inoue and
Rekharsky.1b Data for our CB7−guest pairs appear consistently
within the set of the previously published systems which are
known for their ability to overcome the usual enthalpy−
entropy compensation pattern. In comparison to the previously
published β-CD data set, the lower absolute value of the slope
α and larger intercept −TΔS0 indicate lower influence of
conformational changes and the greater desolvation, respec-
tively.
Titration Calorimetry. Because the determination of
binding affinities of the new guests toward CB7 and β-CD
was the initial intention of this work, we employed titration
calorimetry (ITC) to obtain detailed thermodynamic in-
formation regarding studied systems.
Recently, we reported binding constants for two imidazo-
lium-based guests with CB7.10 The first guest with a short
methylene linker between the adamantane cage and
imidazolium moiety displayed a binding constant K of 3.68
× 1012 M−1. In contrast, the second guest with a longer linker
(oxo-p-xylylene) displayed K = 2.69 × 108 M−1. We also
demonstrated that the contribution of the ion−dipole
interaction to the complex stabilization in the case of the
latter guest is very weak if any. The highest affinity of singly
charged adamantane derivative toward CB7 was reported for 1-
adamantylmethylammonium (K = 7.58 × 1014 M−1),20 and
other common derivatives carrying a single positive charge
displayed K in the order of 1012 M−1 (AdN+Me3I− K = 1.71 ×
1012 M−1, AdN+H3Cl− K = 4.23 × 1012 M−1, AdPy+Br− K =
1.98 × 1012 M−1).4b We assumed that linkers shortened by one
(guests 5a and 5b) or two methylene bridges (guests 10−12)
in comparison to the above-mentioned guest with the oxo-p-
xylylene linker allow more efficient involvement of the ion−
dipole interaction to provide guests with intermediate binding
strengths toward CB7. The obtained results are summarized in
Table 1 (see Figures S34−S38 for detailed titration data). All
five examined guests show a binding strength toward CB7 on
the order of 1010 M−1. Small differences in affinities correlate
with the distance between the cationic moiety and the
adamantane cage. Thus, the guest 5a with the most extended
spacer displays the lowest K value, whereas guests 11 and 12
with cationic moiety directly bound to the phenyl ring display
the highest affinities toward CB7. The correlation of binding
affinity (K) vs linker length between adamantane and
ammonium moiety is shown in Figure S39.
Mass Spectrometry. The ability of all prepared guests to
form supramolecular complexes with β-CD and CB7 was also
studied using mass spectrometry. The equimolar mixtures of
the guest and host and 1:3 mixture of guest 12 and β-CD were
prepared immediately before each measurement. In the first-
order mass spectra of all studied mixtures, the signals, which
can be attributed to complexes with 1:1 stoichiometry, were
observed. These signals were accompanied by signals of singly
charged ions related to the corresponding guests. Additionally,
adducts of singly charged sodium and potassium with β-CD
and doubly charged calcium adduct with two CB7 were also
4485
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
observed in the spectra (Figures S51−S60). The gas-phase
behavior of the complexes was studied using collision-induced
dissociation (CID) after isolation of the required ion.
Nevertheless, the fragmentation was successful only in the
case of mixtures of guests 5a, 5b, and 11 with CB7. As can be
seen from MS/MS spectra, the gas-phase behavior of
complexes 5a@CB7 and 5b@CB7 was similar (Figures S56
and S57). Under the CID conditions, the neutral loss of 1-
methylimidazole from the guest molecule led to the formation
of the singly charged ion observed at m/z 1387, which can be
rationalized as a corresponding benzyl cation complexed with
CB7. Moreover, singly charged CB7 without a single hydrogen
atom (m/z 1161) was observed in both tandem mass spectra
as the most abundant product ion. The suggested fragmenta-
tion pathway is given in Scheme S1. The ability of CB7 to act
as a hydride donor in the reaction with similar benzyl cations is
described in detail in our previous work.21
located inside the β-CD cavity were broadened as well (Figure
1, Figure S22).
Subsequently, we analyzed other titration data to observe the
same pattern. In the case of all examined guests, signals of H
atom at the phenylene ring adjacent to the adamantane cage
and the H atom in the axial position on the unsubstituted
cyclohexane ring of the adamantane cage were broadened
(Figure 2). Because this behavior was observed for all five new
NMR Results. Simultaneous to ITC and MS measure-
1
ments, we performed H NMR titrations to obtain additional
information about complex formation. All examined guests
bind CB7 in a slow-exchange mode as monitored by NMR
spectroscopy. As can be seen in Figures S27−S31, a new subset
of signals for adamantane H atoms appeared at lower values of
a chemical shift, whereas original signals diminished when the
molar fraction of CB7 exceeded 0.5. In concert with the ITC
results, this can be explained by the formation of 1:1 complexes
with the adamantane cage encapsulated into the CB7 cavity.
Because preliminary experiments indicate fast complexation
regime in the case of β-CD, we confirmed 1:1 stoichiometry of
the complexes by Job’s approach22 prior to titration experi-
ments (see Figures S32 and S33). Titrating our guests with β-
CD, we observed the most significant changes in the NMR
resonance area of the adamantane signals. As these signals
moved to higher frequencies, we infer that the adamantane
cage was positioned inside the β-CD cavity in all cases. Results
are given in Figures S19−S23. However, we observed
unexpected selective broadening of some signals during the
experiments with β-CD. The titration experiment of 11 with β-
CD is shown in Figure 1 as an example. The signals of Hd and
Hax became significantly broad when the amount of β-CD in
the mixture exceeded one equivalent (compare 0.8 and 1.2
equiv in Figure 1). Note that signals of the host H atoms
1
Figure 2. Portions of H NMR spectra of 2:3 mixtures of the guests
and β-CD (D2O, 303 K). Signals of phenyl hydrogen atoms in the
ortho positions to 1-adamantyl are asterisked. Signals of adamantane
H atoms in the axial positions (see Figure 1) are assigned with ‡.
guests, we decided to examine this phenomenon in more
detail. One of our initial hypotheses was that there is an
equilibrium of at least two distinct supramolecular arrange-
ments in a moderate-exchange manner. Therefore, we
1
recorded H NMR spectra at various temperatures. While
increased temperature caused only insignificant sharpening of
the signals (spectrum of 5a@β-CD recorded at 323 K is shown
in Figure S19, top line), we clearly observed the splitting of the
guest Hd signal in the spectrum of an equimolar mixture of 11
and β-CD at 278 K (Figure 3, line ii). However, these signals
were still too broad, and other signals were not resolved at all.
Therefore, we tested other approaches to allow recording
spectra at a lower temperature. These tests were performed
1
Figure 3. Portions of H NMR spectra of a 1:1 mixture of guest 11
and β-CD: (i) D2O, 303 K, 500 MHz; (ii) D2O, 278 K, 700 MHz;
(iii) 30% CaCl2 in D2O, 303 K, 700 MHz; (iv) 30% CaCl2 in D2O,
273 K, 950 MHz, the red and blue color indicates the NP and NS
complex, respectively (for representative supramolecular structures of
host−guest complexes, see Figure 5).
1
Figure 1. Portions of H NMR spectra recorded during titration of
the guest 11 (bottom line) by β-CD (D2O, 303 K).
4486
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
1
only with the guest 11 due to its simple H NMR spectrum.
Initially, we tried mixtures of water and DMSO. It is well-
known that DMSO/water mixtures display extreme deviations
from additivity, and a very low freezing point 133 K was
measured at DMSO molar fraction of 0.33.23 Unfortunately,
we did not find any portion of DMSO, which would allow for
both the measurements below 273 K and the observation of
the two resolved sets of signals.
1
Subsequently, we recorded the H NMR spectra in a 30%
solution of CaCl2 in D2O with 5 K steps. Even at 303 K
(Figure 3, line iii), better separation of signals than that in pure
D2O at 278 K was observed. Additional decreasing of the
temperature further improved the separation of all guest
signals. We observed the sharpest peaks and best signal
separation at 273 K. Subsequent decrease of the temperature
led to slight signal broadening, while no progress in signal
separation was observed. Finally, we measured the same
mixture at 950 MHz at 273 K to achieve the perfect resolution
of the two sets of signals (Figure 3, line iv).
Thus, we conclude that these low-temperature NMR results
strongly support the presence of two distinct geometrical
arrangements of the complexes of our new guests with β-CD.
We suggest that two binding arrangements differ in the
orientation of 11 inside of the β-CD cavity. In the first
orientation, denoted as NP, the ammonium part (with
nitrogen [N]) of 11 points toward the primary (P) rim of β-
CD, whereas in the second one, denoted as NS, the
ammonium part aims at the secondary (S) rim. The NS/NP
ratio of 40/60 was estimated at 273 K using the integration of
the spectrum iv in Figure 3. Assignment of NMR signals of
suggested binding forms based on molecular modeling and
1H−1H NOESY experiments will be thoroughly discussed in
the following sections.
Figure 4. (A) Calculated free energy surface as a function of two
collective variables dODIS and αPVANG (Figure S61). Suggested binding
modes correspond to the free energy minima NP and NS, which are
connected by two minimum free energy pathways. IP is an
intermediate (local free energy minimum), and T1P, T2P, and T1S
are transition states (saddle points) along the pathways. Isolines are
spaced by 5 kJ·mol−1. (B) Minimum free energy pathways extracted
from the free energy surface representing interconversion of NP to
NS. Path P shows interconversion occurring at the side of the primary
portal, while path S is at the secondary portal. IP is an intermediate on
path P, and T1P, T2P, and T1S are transition states (see Table S8 for
their summary). ξ is a dimensionless parameter describing positions
along the paths. Confidence intervals (errors) of calculated free
energies are shown as light color strips at three standard deviations.
We also estimated the experimental value of rate constant
and activation free energy barrier for the NS↔NP
interconversion of the 11@β-CD complex in 30% CaCl2
from temperature-dependent measurements at 500 MHz
NMR spectrometer. If we neglect slightly different populations
of both states, the rate constant of interconversion can be
determined approximately in the intermediate exchange range
at the coalescence temperature.24 The coalescence temperature
was determined from a measurement in which two separate
peaks merged into a flat-topped peak. The coalescence
occurred at 350 and 310 K (with a resolution of about 5 K)
for well-resolved aromatic Hd and He protons of guest 11,
respectively. Observed separation of Hd and He signals in the
slow-exchange regime was 0.33 and 0.04 ppm, which provided
the following rate constants: 367 s−1 at 350 K and 44 s−1 at 310
K. Finally, the activation free energy barriers obtained by
applying the Eyring equation were determined as 69.0 and 66.2
kJ·mol−1, respectively.
Molecular Dynamic Simulation of 11@β-CD. NMR
observations suggested that the time scale for interconversion
between two forms of 11@β-CD complex (∼ms) is beyond
the possibilities of regular molecular dynamic (MD)
simulations (∼μs). Thus, we employed biased adaptive biasing
force/multiple walker (ABF/MWA) simulations, allowing us
to sample all possible arrangements of 11 and β-CD in a
reasonable computational time (see the Computational
Methods and Supporting Information). Using two collective
variables (Figure S61) for the description of the mutual
position (dODIS) and orientation (αPVANG) of 11 and β-CD, we
obtained the free energy surface shown in Figure 4A.
Two suggested binding modes, NP and NS, appeared as
local minima on the calculated surface. The calculation
indicated that the state NP could not be simply converted to
NS through rotation of 11 inside the cavity, as this process
would result in the enormous rise of the free energy (see the
red part around dODIS ≈ 0 Å in Figure 4A). Instead, there are
two possible interconversions represented as two minimum
free energy pathways connecting NP and NS (Figure 4B, Table
S8). While 11 dissociates on the side of the primary rim along
path P, it leaves the cavity at the side of the secondary rim
along path S. Additionally, path P features one extra free
energy minimum (intermediate IP). Representative structures
belonging to important thermodynamic states were extracted
from ABF/MWA trajectories and are depicted in Figure 5. In
accordance with 1H NMR experiments, both NP and NS have
the adamantane cage inside the cavity, while the rest of the
guest is located outside the macrocycle. In the case of NP, the
phenyltrimethylammonium arm is on the primary side of β-
CD, while it is on the secondary side of β-CD in NS. In
addition, we found an intermediate IP, which captures 11
bound outside the β-CD cavity. The structure of IP shows a
parallel alignment of 11 with the β-CD plane and interaction
with the primary portal of β-CD.
The obtained free energy surface exhibits only a qualitative
thermodynamic agreement with the experiment. The calcu-
lation suggests that NS is more stable than NP (ΔGNS−NP
=
−5.1 1.5 kJ·mol−1 at 300 K, c(NaCl) = 0.14 M), while NMR
showed it the opposite (ΔGNS−NP = +1.2 kJ·mol−1 at 273 K,
c(CaCl2) ≈ 3.4 M). However, the difference between NS and
NP was found to be small in both cases, which indicates that
4487
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
is 0.002% at 300 K, indicating that IP would be barely
detectable by NMR. The size of barriers along both pathways
suggests that path S is likely to be the preferred way of the
interconversion with a rate-limiting step of 41.0 kJ·mol−1,
compared to path P with the highest transition state at 51.9 kJ·
mol−1. Barriers obtained from the modeling are under-
estimated in comparison to the experimentally determined
value (∼68 kJ·mol−1). Apart from the inaccuracies of the
employed empirical force field, the most significant impact on
the barrier seems to be caused by high salt concentration under
experimental conditions (i.e., 30% CaCl2). This behavior is
also corroborated experimentally since the poor signal
separation of NP and NS forms were observed in pure D2O
solution even at temperatures close to 273 K.
Selection of right interconversion pathway between the NP
and NS state is not critical for drawn conclusions as the
calculated free energy is a state function, and thus, the
calculated difference between NS and NP states is path
independent (within numerical accuracy). In the case of the
guest 11, we have shown that interconversion can take place by
threading the guest through β-CD cavity. However, this does
not exclude other possibilities including partial or full complex
dissociation, which can be preferred by other guests (5a, 5b,
10, and 12) showing different bulkiness of the cationic part.
Experimentally, the interconversion by the full complex
dissociation is not excluded even for 11 as the combination
of ITC binding affinity (ΔGb=−34.2 kJ·mol−1) and kinetic data
Figure 5. Representative structures along two possible interconver-
sion pathways from NP to NS. Guest 11 is represented in dark gray,
β-CD in light gray, O atoms in red, N atoms in blue.
two states would be similarly populated, as observed
experimentally. The difference is most likely caused by
inaccurate force field parameters and different environmental
conditions than in the experiment (low temperature and high
salt concentration).
Most importantly, no stable complex arrangements other
than NP and NS were detected by extensive ABF/MWA
sampling. The estimated population of IP from calculated data
Figure 6. Portions of the 2D NOE spectrum (top) recorded on a 1:1 mixture of guest 11 and β-CD in 30% CaCl2 in D2O at 273 K at two mixing
times. Shorter mixing time (50 ms) provided information about close H−H contacts (up to 3 Å), whereas a longer mixing time (80 ms) provided
information about longer H−H contacts (up to 6 Å). Horizontal lines represent H5′ and H3′ signals for NP (solid) and NS (dashed) forms.
Idealized geometries of NP and NS forms with annotations of hydrogen atoms and contacts (bottom). Corresponding intermolecular dipolar
contacts in the NOESY spectrum and idealized structures are coded by matching colors.
4488
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(ΔGa = 66.2 kJ·mol−1) from NMR measurements indicates
that the transition state for interconversion is above the
dissociated state.
Linking Calculated Structures with NMR Data. 1H
NMR experiments and molecular modeling suggested that
there are two stable arrangements of the complex. Whereas the
adamantane cage occupies the interior of β-CD cavity in both
complexes, the phenyltrimethylammonium moiety is protrud-
ing either from the primary (P) or secondary (S) cyclodextrin
portal (Figure 5). We performed a 2D NOE experiment25 on
the sample related to line iv in Figure 3 to assign the NMR
signal subsets to the corresponding supramolecular arrange-
ments.
1
We found a different pattern of dipolar H−1H contacts
between NP and NS binding modes (Figure 6). While contacts
between phenyl ring and β-CD were detected for NP (H5′···
Hd, and H5′···He), analogous contacts with H5′ or H3′ were
not observed for NS. This qualitatively agrees with the interior
arrangement of 11 inside the β-CD cavity. In both orientations,
the adamantane cage prefers a position of about 1.0 Å off a
plane going through β-CD glycosidic bridges toward the
secondary rim (Figure S62). Such asymmetry moves the
phenyl ring closer to interior protons of β-CD in NP but
farther in NS.
For the full assignment of NOESY data, we calculated
proton contacts from unbiased molecular dynamics simulations
of NP and NS states (Figure 7). Most of the close contacts,
which exhibited a higher occurrence for shorter distances, were
also observed in the NOESY spectrum. However, we found a
few exceptions. H3′···Hax shows small occurrence for both
forms, and only NP H3′···Hax was detected in NOESY at a
longer mixing time. Next, NOESY did not detect any contact
of H3′ with either aromatic protons Hd and He, although RDF
showed high occurrences for NS H3′···Hd. We assume that the
side arm of the guest containing the phenyl ring, which is
outside of β-CD, is wiggling fast, especially in the NS state,
resulting in an NOE peak broadening and its disappearance
from the NMR spectrum. This behavior is also supported by
the free energy surface, which shows a larger basin prolonged
in αPVANG dimension for NS compared to NP.
Figure 7. Radial distribution occurrence for contacts between proton
of guest 11 and the H5′ (left) and H3′ (right) protons of β-CD from
unbiased molecular dynamics of NP and NS orientations. Detected
(√) and undetected (×) dipolar contacts in the NOESY spectrum
are color coded in the same way as in Figure 6.
In addition to 2D NOE experiments, we analyzed trends in
complexation-induced NMR shifts. NMR data indicate that
H5′ of β-CD in the NP form is more shielded compared to NS
and ΔδNHP−NS=−0.33 ppm, respectively. We assume that the
NP−NS
(Δδ
= −0.18 ppm), which can be attributed to the ring-
H5′
d
current through-space effects of the phenyl ring being closer to
H5′ in structure. An opposite but smaller difference in the
changes are induced mainly by desolvation of hydrogen atoms
upon encapsulation and the shielding effect of the cavity
interior of β-CD. Therefore, the effect of desolvation was
evaluated in the next step. The radial distribution function for
ligand···water calculated from MD simulations (Figure S63)
clearly shows that Hax is substantially more accessible to
solvent molecules in the NP orientation, whereas the Hc and
Hd atoms are exposed more in the NS orientation.
To capture both desolvation and the shielding effect, we
attempted to calculate NMR shifts of ligand H atoms as a
function of ligand orientation (NP/NS) and axial distance
between the ligand and center of mass of β-CD. The NMR
shielding was calculated by employing the DFT approach in an
implicit water model applied on a series of structures differing
just in ligand···host separation (for details, see the Supporting
Information). An obtained scan depicted in Figure 8
corroborates that Hax in the NS shows lower chemical shifts
compared to the NP orientation. Furthermore, Hd and Hc
exhibit significantly lower chemical shifts in the NP since they
NMR chemical shift was also observed for the H3′ signal
NP−NS
(Δδ
= +0.08 ppm). For comparison, we calculated the
H3′
through-space shielding (as nucleus-independent chemical
shifts [NICS]) on a regular grid around the benzene molecule.
These data were superimposed on the phenyl ring in each
snapshot of unbiased molecular simulations of NP and NS
states to obtain an average effect on the host protons.10
NP−NS
H5′
NP−NS
= +0.19
H3′
Obtained data Δδ
= −0.17 ppm and Δδ
ppm show good qualitative agreement with experimental data,
confirming that structures observed in molecular simulations
are similar to those observed experimentally.
1
We have also compared H NMR chemical shifts of ligand
atoms between the binding modes. The largest relative
perturbation was observed for the terminal Hax atoms, which
are located on one side of β-CD, and the Hc and Hd protons,
which are located on the opposite side of β-CD, with chemical
shift changes of ΔδNHP−NS = +0.21 ppm, ΔδNHP−NS = −0.12 ppm
ax
c
4489
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
1
Figure 8. Rigid body scan shows changes of H relative chemical shift of the ligand 11 upon encapsulation into the β-CD cavity through the
primary (left) and secondary (right) portal calculated by the DFT approach in an implicit water solvent. Resulting experimental differences in
NMR chemical shifts between NP and NS orientation ΔδNP−NS are summarized in the table inset. Data shown in the CAL column refer to
differences in NMR shifts predicted by the DFT for mutual geometries (red and blue stripes in the graphs) corresponding to unbiased molecular
dynamics for NP and NS (Figure S62).
are exposed to the narrower primary portal. Despite the
limitation of the employed model, which does not account for
geometry relaxation or tilting of the guest upon complexation,
the analysis provided good agreement between experimental
and calculated changes of chemical shifts of NP and NS forms
(see the table inset in Figure 8).
them, we demonstrated that ion−dipole interactions contrib-
ute to the stabilization of the complex with CB7 as very little
for the former and significantly for the latter. In this work, we
confirmed this trend. All of the guests with the phenyl spacer
displayed intermediate values of association constants toward
CB7 (K ≈ 1010 M−1). In other words, the stability of the
complex negatively correlates with the length of the linker
between the adamantane cage and the cationic moiety as the
contribution of the ion−dipole interactions decreases with the
length of the linker.
According to the ITC experiments, our new guests showed
surprisingly high association constants with β-CD within the
range of (0.6−2.6) × 106 M−1. The highest value (Kβ‑CD = 2.6
× 106 M−1), which was obtained for guest 5a, rivals the highest
binding constant ever reported for a 1:1 β-CD complex
(triamantane-9-carboxylic acid, Kβ‑CD = 2.9 × 106 M−1).3 We
suggest our ligand 5a as a reasonable choice for the design of
multitopic guests because the overall yield of the correspond-
ing building block 4a from commercially available 1-
bromoadamantane was 30%, whereas the yield of triaman-
tane-9-carboxylic acid was 19% from a rather uncommon
triamantane.27 1H NMR experiments on complexes of β-CD
with our guests revealed a moderate-exchange process in the
mixtures containing an excess of β-CD in water. We assumed
that two distinct geometrical forms of the guest@β-CD
complex were present in equilibrium. This hypothesis was
strongly supported by 1H NMR experiments on the mixture of
guest 11 and β-CD. We performed these measurements in 30%
CaCl2 solution in D2O to be able to record spectra within the
temperature range of 258−303 K. We observed two
CONCLUSIONS
■
In concert with our original motivation, which was focused on
the preparation of suitable binding motifs for CB7 with
moderate affinity, we synthesized five new model guests and
determined their binding properties toward CB7 and β-CD.
The guests consist of 1-adamantylphenyl scaffold and various
cationic moiety, namely ammonium (11), imidazolium (5a,
5b, 10), and pyridinium (12). We prepared only methylated
guests (with exception to pyridinium salt 12), but the
employed procedure is general and its extension to other
alkylating agents is feasible. In particular, imidazolium
derivatives can serve for the construction of advanced
supramolecular components such as multitopic guests.26
Combining data from ITC and NMR, we found that all
examined guests bind CB7 with an adamantane cage
positioned inside the CB7 cavity. The values of association
constants KCB7 were determined within the range (0.6−5) ×
1010 M−1. These values match our expectations because they
lie nearly in the middle of the interval given by association
constants of previously reported guests10 of similar structure:
4-(1-adamantylcarbonyl)phenylmethyl(methyl)imidazolium
iodide (KCB7 = 2.6 × 108 M−1) and 1-adamantylmethyl-
(methyl)imidazolium iodide (KCB7 = 3.7 × 1012 M−1). For
4490
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Electrospray mass spectra (ESI-MS) were recorded using an amaZon
X ion-trap mass spectrometer (Bruker Daltonics, Bremen, Germany)
equipped with an electrospray ionization source. All the experiments
were conducted in the positive-ion polarity mode. The instrumental
conditions used to measure the single imidazolium salts and their
mixtures with the host molecules were different; therefore, they are
described separately. Single guests: Individual samples (with
concentrations of 0.5 μg·cm−3) were infused into the ESI source in
MeOH:H2O (1:1, v-v) solutions using a syringe pump with a constant
flow rate of 3 μL·min−1. The other instrumental conditions were as
follows: an electrospray voltage of −4.2 kV, a capillary exit voltage of
140 V, a drying gas temperature of 220 °C, a drying gas flow rate of
6.0 dm3·min−1 and a nebulizer pressure of 55.16 kPa. Host−guest
complexes: An aqueous solution of the guest (12.5 μM) and the
equimolar amount of the corresponding host (in the case of 12@β-
CD 3.0 equiv of host were used) was infused into the ESI source at a
constant flow rate of 3 μL·min−1. The other instrumental conditions
were as follows: an electrospray voltage of −4.0 kV, a capillary exit
voltage of 140 V up to −50 V, a drying gas temperature of 300 °C, a
drying gas flow rate of 6.0 dm3·min−1, and a nebulizer pressure of
206.84 kPa. Nitrogen was used as both the nebulizing and drying gas
for all of the experiments. Tandem mass spectra were collected using
CID with He as the collision gas after the isolation of the required
ions. Isothermal titration calorimetry measurements were carried out
in H2O using a VP−ITC MicroCal instrument at 303 K. The
concentrations of the host in the cell and the guest in the
microsyringe were approximately 0.05 mM and 0.50 mM for CB7
and 0.15 mM and 1.50 mM for β-CD, respectively. The raw
experimental data were analyzed with the MicroCal ORIGIN
software. The heats of dilution were taken into account for each
guest compound. The data were fitted to a theoretical titration curve
using the One Set of Sites model. If needed, a competitive approach7
unambiguously resolved subsets of signals for the guest below
273 K. The dipolar contacts observed in 2D NOE experiments
allowed us to identify two geometrical arrangements where
phenyltrimethylammonium moiety protrudes either from the
primary (NP) or secondary (NS) portal of β-CD. According to
1H NMR spectrum, the NP:NS ratio was 60:40 in 30% CaCl2
at 273 K. Finally, two different binding modes of guest 11
inside the β-CD cavity were confirmed by the unbiased and
biased MD simulations. Calculated free energy shows that the
NP and NS arrangements are of comparable free energies. We
also described two pathways for their interconversion, which
both took place through partial complex dissociation. The rate-
determining step had barrier 41 kJ·mol−1 under simulation
conditions (NaCl solution), which is close to the exper-
imentally determined value of 68 kJ·mol−1 (CaCl2 solution).
In this work, we presented convenient synthetic approaches
leading to new guests, which are suitable for binding into β-CD
and CB7 macrocycles with tunable affinity. The structural
characterization of such supramolecular complexes, which is
essential for further rational design, is challenging due to its
inability to obtain single crystals suitable for the X-ray
diffraction analysis. Therefore, we combined low-temperature
NMR in salt water and computer simulations to obtain
comprehensive information about their structure. We demon-
strated that this state-of-the-art methodology is able to throw
light on intrinsic structure of cyclodextrin complexes. Despite
the long history of the cyclodextrin chemistry, actual
orientation of the guests in the conical CD cavity has not
been studied thoroughly, most likely due to fast equilibration
of relatively weak complexes. As the particular forms can differ
in their binding strengths, detailed knowledge of the complex
geometry is especially important if only one rim of the
cyclodextrin macrocycle can be employed for host−guest
interactions in considered systems. We believe that our
methodology is suitable for the study of functional supra-
molecules, which can act as drug delivery systems, molecular
sensors or switches.
was employed using a 1-methyl-3-hexylimidazolium chloride (KCB7
=
1.23 × 107 dm3·mol−1) as a competitor. The K values obtained from
the competitive titrations were verified using two different
concentrations of a competitor. All titrations were performed in
triplicate.
X-ray Crystallographic Analysis of 3b and 4b. The crystals of
3b were grown in the mixture of 3b/3a after column chromatography
at room temperature. The single crystals were collected by tweezers
and washed using EtOAc. An identical crystal structure has been very
recently reported.34 The crystals of 5b were grown by slow
evaporation of D2O at room temperature. Single-crystal X-ray
diffraction data were collected on a Rigaku MicroMax-007 HF
rotating anode four-circle diffractometer using Mo Kα radiation at
120 K. The non-hydrogen atoms were refined anisotropically,
hydrogen atoms were refined as riding on their carrier atoms.
CrystalClear and CrysAlisPro software packages were used for data
collection and reduction.35 The structures were solved by the direct
methods procedure and refined by full matrix least-squares methods
on F2 using SHELXT and SHELXL.36 The crystallographic data of 3b
and 5b were deposited in the Cambridge Crystallographic Data
Centre with the deposition numbers CCDC 1956556 and CCDC
1956557, respectively. For further details of XRD measurements, see
the Supporting Information.
Computational Methods. Two complexes of 11@β-CD were
built by placing 11 into the cavity of β-CD in two possible
arrangements NP and NS. Each complex was separately immersed
into a truncated octahedral box filled with TIP3P water and sodium
and chloride ions (cNaCl ≈ 0.14 M). General Amber Force Field 237
was utilized for 11, whereas β-CD was described by GLYCAM06.38
Unbiased MD simulations were performed in the Amber 16.0
package39 at 300 K and 100 kPa and were 1 μs long each. The
interconversion between NP and NS was further studied by biased
MD using the ABF method40 enhanced by the MWA approach41 in
the modified pmemd program from the Amber package coupled with
PMFLib.42 The total length of the biased ABF/MWA sampling was
4.4 μs. The resulting free energy surface was reconstructed from the
calculated mean forces by the Gaussian Process Regression (GPR)43
EXPERIMENTAL SECTION
■
General Information. All solvents, reagents, and starting
compounds were of analytical grade, purchased from commercial
sources, and used without further purification, if not stated otherwise.
Tolyladamantanes 2a and 2b,28 bromo derivatives 3a and 3b,29 1-
phenyladamantane (6),30 nitro derivative 7,31 and aniline 832 were
prepared following previously published procedures. Melting points
were measured on a Kofler block and are uncorrected. Elemental
analyses (C, H and N) were performed using a Thermo Fisher
Scientific Flash EA 1112. HRMS analyses were performed on an
Agilent 6230 Time-of-Flight spectrometer with an electrospray ion
source. NMR spectra were recorded using a Bruker Avance NEO 500
MHz spectrometer operating at frequencies of 500.21 MHz (1H) and
125.78 MHz (13C), a Bruker Avance III HD 700 MHz spectrometer
operating at a frequency of 700.80 MHz (1H), a Bruker Avance III
HD 950 MHz spectrometer operating at a frequency of 950.33 MHz
(1H) and a Jeol JNM-ECZ400R/S3 spectrometer operating at
1
frequencies of 399.78 MHz (1H) and 100.53 MHz (13C). H and
13C NMR chemical shifts were reported as parts per million (ppm)
and referenced to the signal of the solvent (1H: δ[residual DMSO-d5]
= 2.50 ppm, δ[residual HDO] = 4.70 ppm, δ[residual CHCl3] = 7.27
ppm; 13C: δ[DMSO-d6] = 39.52 ppm, δ[CDCl3] = 77.16 ppm). The
25,33
1
mixing time for 2D H−1H NOESY experiment
was adjusted to
50 or 80 ms. Signal multiplicity is indicated by “s” for singlet, “d” for
doublet, “m” for multiplet, and “um” for unresolved multiplet. IR
spectra were recorded using a Smart OMNI-Transmission Nicolet
iS10 spectrophotometer. Samples were measured in KBr pellets.
4491
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
with optimized GPR hyperparameters. All errors are reported at three
standard deviations.
organic portions were washed with brine (15 mL) and dried over
sodium sulfate, and the solvents were evaporated under reduced
pressure to obtain a colorless powder. The crude product was purified
by column chromatography (silica gel, petroleum ether) to obtain a
colorless microcrystalline powder (0.1792 g, 0.84 mmol, yield 87%).
Mp: 89−90 °C (lit.44 mp 87−89 °C). 1H and 13C NMR data
correspond to that previously published.45
Snapshots from the unbiased MD were also used to investigate how
shielding of the benzene ring found in 11 propagates onto the protons
of β-CD via the NICS approach.10 Changes in chemical shifts were
calculated at B3LYP/def2-SVP(β-CD), def2-TZVPP(ligand)/SMD
level of theory. An extended description of the computational
methodology is provided in Supporting Information.
1-(4-Nitrophenyl)adamantane (7) was prepared following the
previously described procedure.31 Starting acetyl nitrate was prepared
from Ac2O (23 mL) and a mixture of concd HNO3 (9.3 mL) and
concd H2SO4 (0.45 mL) in a double-coated reaction flask at −15 °C.
Subsequently, a dispersion of 6 (2.005 g, 9.44 mmol) in acetic
anhydride (40 mL) was slowly added dropwise while the internal
temperature of the reaction mixture was maintained within the range
of −15 to −5 °C. When the compound 6 was consumed (according
to TLC or GC), the reaction was quenched by addition of crushed ice
to precipitate a pale yellow solid. The water dispersion was extracted
with several portions of Et2O, and the collected organic portions were
treated with water (2 × 50 mL) and saturated NaHCO3 solution and
dried over sodium sulfate. The solvent was evaporated, and the crude
product was crystallized from MeOH to obtain yellowish needles
(1.3851 g, 5.38 mmol, yield 57%). Mp: 128−130 °C (lit.46 mp: 129−
Synthesis. 1-Tolyladamantanes 2a and 2b. Method A: The
reaction was carried out according to a slightly modified published
procedure.28 The catalyst InCl3·H2O (0.0196 g, 0.082 mmol) was
dried prior to the reaction by refluxing with SOCl2 (5 cm3) which was
then evaporated. To a dry catalyst, compound 1 (0.3611 g, 1.68
mmol) and dry toluene (5 cm3) were added. The reaction mixture
was stirred under an inert atmosphere at 30 °C (maintained using an
oil bath) until GC−MS indicated consumption of the compound 1.
The reaction was quenched with 115 cm3 of 10% NaHCO3 and
mixture was extracted with AcOEt (3 × 10 cm3). Combined organic
portions were washed with brine (15 cm3) and dried over anhydrous
sodium sulfate. Solvents were evaporated under reduced pressure and
the crude product was purified by column chromatography (silica gel,
petroleum ether) to isolate a mixture of 2a and 2b as a colorless
microcrystalline powder (0.3529 g, 1.56 mmol, yield 93%, 2a:2b =
19:1 according to GC−MS).
Method B: Compound 1 (2.0463 g, 9.51 mmol) was added to a
dispersion of AlCl3 (0.0450 g, 0.34 mmol) in dry toluene (25 cm3).
The reaction mixture was stirred under an inert atmosphere at −3 °C
(cooled by ice/water/NaCl mixture) and monitored using GC−MS.
Since the compound 1 was completely consumed, the reaction was
quenched with 10% NaHCO3 solution (15 cm3) and extracted with
AcOEt (3 × 10 cm3). Combined organic portions were washed with
brine (15 cm3), dried over sodium sulfate and the solvents were
evaporated under reduced pressure to obtain a yellowish powder. The
crude product was purified by column chromatography (silica gel,
petroleum ether) to isolate a mixture of 2a and 2b as a colorless
microcrystalline powder (1.1560 g, 5.11 mmol, yield 54%, 2a:2b = 1:2
according to GC−MS). MS (EI, 70 eV): 2a 41 (12); 65 (7); 77 (11);
79 (13); 91 (20); 94 (13); 105 (21); 115 (13); 133 (7); 141 (7); 153
(5); 154 (10); 155 (6); 169 (100); 170 (19); 183 (12); 226 (73);
227 (14) m/z (%); 2b 41 (14); 65 (7); 67 (5); 77 (13); 79 (17); 91
(22); 940 (27); 105 (21); 115 (13); 117 (6); 128 (11); 132 (15);
141 (9); 153 (5); 154 (10); 155 (7); 169 (100); 170 (21); 183 (14);
226 (77); 227 (15) m/z (%).
1-(Bromomethylphenyl)adamantanes 3a and 3b were prepared
following the previously described procedure.29 Typically, N-
bromosuccinimide (0.2514 g, 1.41 mmol) was added to a solution
of 2a and 2b mixture (0.3070 g, 1.36 mmol, 2a:2b = 19:1) in dry
CCl4 (5 mL). The reaction mixture was refluxed under an inert
atmosphere using an oil bath and irradiated by 60 W tungsten lamp
for 1 h. After the mixture was cooled to room temperature,
succinimide was filtered off on a sintered-glass funnel and solvent
was evaporated under reduced pressure to obtain a yellow oil. The
crude product was purified by column chromatography (silica gel,
petroleum ether) to isolate a mixture of 3a and 3b as a yellowish
powder (0.2961 g, 0.97 mmol, yield 71%, 3a:3b = 19:1 according to
GC−MS). The same procedure was repeated with the 1:2 mixture of
2a:2b to yield oily product from which 2b slowly crystallized at room
temperature. MS (EI, 70 eV): 3a 41 (6), 77 (5), 79 (6), 91 (13), 104
(5), 105 (9), 115 (8), 131 (11), 167 (6), 168 (6), 169 (10), 225
(100), 226 (24), 304 (2), 306 (2) m/z (%); 3b 77 (7), 79 (12), 91
(15), 93 (5), 94 (5), 128 (8), 129 (6), 141 (6), 167 (10), 169 (13),
225 (100), 226 (27), 304 (4), 306 (4) m/z (%).
1-Phenyladamantane (6) was prepared following the previously
described procedure.30 Compound 1 (0.2093 g, 0.97 mmol) was
added to a dispersion of AlCl3 (0.0293 g, 0.22 mmol) in dry benzene
(10 mL). The reaction mixture was stirred under an inert atmosphere
at room temperature, and the reaction progress was monitored using
GC−MS. When GC indicated the complete consumption of the
compound 1, the reaction was quenched with 10% NaHCO3 solution
(10 mL) and extracted with AcOEt (3 × 10 mL). The combined
1
130 °C). H NMR (CDCl3, 500 MHz): δ 8.17 (d, 2H, J = 8.4 Hz),
7.51 (d, 2H, J = 8.4 Hz), 2.15 (m, 3H), 1.94 (m, 6H), 1.76−1.85 (m,
6H).
4-(1-Adamantyl)aniline (8) was prepared following a slightly
modified procedure reported previously.32 Compound 7 (3.2153 g,
12.50 mmol) was dissolved in MeOH (250 mL), and concd HCl (36
mL) was added. Portions of iron powder (approximately 1.65 g, 30
mmol) were added successively into the refluxed mixture (using an oil
bath) and stirred unless TLC indicated the consumption of 7. The pH
of the reaction mixture was controlled and kept acidic by addition of
small portions of concd HCl. After five portions of iron were added,
the mixture was poured to 15% NaOH solution (150 mL) and
extracted several times with Et2O. The collected organic portions
were washed with brine, dried over sodium sulfate, and concentrated
under reduced pressure to obtain brown solid (2.1813 g, 9.60 mmol,
yield 77%). This crude material was used in further steps without
purification. Mp: 90−92 °C (lit.47 mp: 98−101 °C). 1H NMR
(DMSO-d6, 400 MHz): δ 6.98 (m, 2H), 6.49 (m, 2H), 4.75 (s, 2H),
2.02 (um, 3H), 1.77 (m, 6H), 1.70 (um, 6H). 13C{1H} NMR
(DMSO-d6, 101 MHz): δ 146.0, 138.4, 124.8, 113.7, 43.0, 36.3, 34.7,
28.4. MS (EI, 70 eV): 227 (70), 170 (100), 152 (3), 133 (30), 115
(5), 91 (12), 77(9) m/z (%).
1-(4-(1-Adamantyl)benzyl)-1H-imidazole (4a) was prepared
following the previously described procedure.10,21 The commercial
NaH (60%, 0.0597 g, 1.49 mmol) was washed with dry pentane under
an inert atmosphere to remove mineral oil. After a dispersion of NaH
in DMF (3 mL) was prepared, a solution of imidazole (0.1006 g, 1.48
mmol) in DMF (5 mL) was slowly added. This reaction mixture was
stirred for 15 min, and the solution of bromides 3a and 3b (0.2981 g,
0.98 mmol, 3a:3b = 19:1) in DMF (3 mL) was added in one portion.
The reaction mixture was heated at 120−130 °C using an oil bath and
stirred until GC−MS indicated the complete consumption of starting
alkyl bromides. After 4 h, the reaction mixture was poured into
crushed ice and extracted with CH2Cl2 (5 × 20 mL). Combined
organic portions were washed with brine (20 mL) and dried over
sodium sulfate. After the removal of the solvent under reduced
pressure, the crude product was purified by repeated column
chromatography (silica gel, CHCl3/AcOEt, 1:1, v/v). After two
runs on the column, a colorless microcrystalline powder of 4a was
obtained (0.1367 g, 0.47 mmol, yield 48%). Mp: 152−154 °C. Anal.
Calcd for C20H24N2: C, 82.15; H, 8.27; N, 9.58. Found: C, 82.03; H,
8.33; N, 9.41. 1H NMR (CDCl3, 500 MHz): δ 8.22 (s, 1H), 7.37 (m,
2H), 7.17 (um, 2H), 6.98 (s, 1H), 5.21 (s, 2H), 2.10 (um, 3H), 1.89
(m, 6H), 1.77 (m, 6H). 13C{1H} NMR (CDCl3, 126 MHz): δ 152.3,
136.4, 131.6, 127.7, 125.9, 125.8, 119.8, 51.6, 43.1, 36.7, 36.2, 28.9. IR
(KBr): 3448 (w), 3105 (sh), 3098 (vw), 3052 (vw), 3023 (vw), 2965
(sh), 2920 (sh), 2905 (vs), 2848 (s), 1507 (w), 1448 (w), 1417 (vw),
1344 (vw), 1320 (vw), 1285 (vw), 1280 (sh), 1262 (vw), 1249 (sh),
4492
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
1237 (sh), 1231 (m), 1231 (m), 1126 (vw), 1112 (sh), 1108 (vw),
1102 (vw), 1080 (w), 1070 (w), 1032 (w), 1016 (w), 976 (vw), 907
(m), 841 (sh), 836 (vw), 811 (w), 800 (m), 758 (w), 735 (m), 710
(sh), 705 (vw), 661 (m), 635(vw),628 (sh), 626 (sh), 598 (vw), 533
(m) cm−1. ESI-MS (pos.) m/z (%): 293.2 [M + H+]+ (100), 585.4 [2·
M + H+]+ (16).
colorless powder (0.0884 g, 0.20 mmol, yield 69%). Mp: 162−164
°C. Anal. Calcd for C21H27IN2·0.8H2O: C, 56.20; H, 6.42; N 6.24.
Found: C, 56.25; H, 6.43; N, 6.22. HRMS (ESI/TOF) m/z: M+
1
Calcd for C21H27N2 307.2169; Found 307.2175. H NMR (CDCl3,
500 MHz): δ 10.37 (s, 1H), 7.40 (s, 4H), 7.29 (s, 1H), 7.21 (s, 1H),
5.49 (s, 2H), 4.09 (s, 3H), 2.09 (um, 3H), 1.88 (m, 6H), 1.76 (m,
6H). 13C{1H} NMR (CDCl3, 126 MHz): δ 153.3, 138.1, 129.6, 129.1,
126.3, 123.5, 122.0, 53.9, 43.2, 37.4, 36.8, 36.5, 29.0. IR (KBr): 3440
(m), 3166 (vw), 3131 (vw), 3096 (vw), 3081 (w), 2964 (w), 2925
(sh), 2910 (sh), 2903 (sh), 2889 (sh), 2877 (s), 2849 (s), 1567 (m),
1561 (sh), 1554 (sh), 1514 (w), 1507 (s), 1457 (sh), 1450 (w), 1420
(sh), 1416 (w), 1409 (vw), 1361 (w), 1356 (sh), 1348 (w), 1342 (w),
1332 (vw), 1321 (vw), 1319 (sh), 1273 (vw), 1160 (sh), 1144 (w),
1107 (w), 1104 (sh), 1101 (sh), 1034 (vw), 1022 (w), 1018 (sh), 976
(vw), 862 (vw), 852 (w), 813 (m), 804 (m), 774 (w), 768 (w), 762
(sh), 758 (w), 745 (sh), 710 (w), 691 (w), 664 (sh), 657 (w), 623
(m), 614 (w), 593 (vw), 539 (m) cm−1. ESI-MS (pos.) m/z (%):
307.1 [M+]+ (100).
1-(3-(1-Adamantyl)benzyl)-1H-imidazole (4b) was prepared
following an analogous procedure as described for 4a. However, the
starting mixture of bromides 3a and 3b was prepared by method B.
This mixture of 3a and 3b (0.4165 g, 1.36 mmol, 3a:3b = 1:2)
provided a colorless microcrystalline powder of 4b (0.080 g, 0.27
mmol, yield 20%). Mp: 136−139 °C. Anal. Calcd for C20H24N2: C,
82.15; H, 8.27; N, 9.58. Found: C, 81.93; H, 8.35; N, 9.48. 1H NMR
(CDCl3, 500 MHz): δ 7.78 (s, 1H), 7.34 (um, 1H), 7.30 (m, 1H),
7.17 (s, 1H), 7.12 (s, 1H), 6.97 (um, 1H), 6.93 (s, 1H), 5.15 (s, 2H),
2.09 (um, 3H), 1.87 (m, 6H), 1.76 (m, 6H). 13C{1H} NMR (CDCl3,
126 MHz): δ 152.6, 137.3, 135.4, 129.0, 128.4, 125.3, 124.8, 124.2,
119.7, 51.7, 43.2, 36.8, 36.4, 29.0. IR (KBr): 3100 (vw), 3024 (vw),
2902 (vs), 2846 (vs), 1606 (w), 1587 (vw), 1504 (m), 1449 (w),
1392 (vw), 1367 (vw), 1355 (vw), 1343 (w), 1318 (w), 1278 (w),
1261 (vw),1225 (w), 1184 (vw), 1169 (vw), 1106 (m), 1090 (m),
1073 (m), 1049 (w), 1027 (m), 975 (w), 905 (w), 865 (vw), 815
(vw), 806 (w), 761 (w), 736 (s), 699 (m), 663 (m), 628 (w), 634
(sh), 547 (vw), 442 (vw) cm−1. ESI-MS (pos.) m/z (%): 293.2 [M +
H+]+ (100), 585.4 [2·M + H+]+ (13).
1-(3-(1-Adamantyl)benzyl)-1H-imidazolium iodide (5b) was
prepared from compound 4b (0.0556 g, 0.19 mmol) to yield a
colorless powder (0.0509 g, 0.12 mmol, yield 63%). Mp: 171−173
°C. Anal. Calcd for C21H27IN2: C, 58.07; H, 6.27; N, 6.45. Found: C,
1
57.94; H, 6.36; N, 6.39. H NMR (DMSO-d6, 500 MHz): δ 9.20 (s,
1H), 7.81 (um, 1H), 7.70 (um, 1H), 7.48 (um, 1H), 7.38 (um, 1H),
7.35 (m, 1H), 7.20 (um, 1H), 5.39 (s, 2H), 3.86 (s, 3H), 2.07 (um,
3H), 1.86 (m, 6H), 1.74 (m, 6H). 13C{1H} NMR (DMSO-d6, 126
MHz): δ 151.8, 136.5, 134.5, 128.7, 125.4, 125.2, 125.0, 124.0, 122.3,
52.2, 42.5, 36.1, 35.9, 35.8, 28.2. IR (KBr): 3418 (vw), 3142 (vw),
3126 (vw), 3076 (vs), 3052 (sh), 2971 (w), 2923 (vs), 2911 (vs),
2895 (vs), 2848 (vs), 1605 (w), 1567 (m), 1489 (w), 1448 (sh), 1430
(w), 1372 (w), 1364 (sh), 1344 (w), 1319 (w), 1273 (vw), 1245
(vw), 1201 (vw), 1171 (w), 1151 (vs), 1103 (vw), 1094 (vw), 980
(vw), 976 (vw), 832 (w), 823 (w), 816 (m), 798 (w), 768 (vw), 745
(s), 707 (m), 698 (w), 666 (w), 624 (m), 616 (s), 592 (vw), 545
(vw) cm−1. ESI-MS (pos.) m/z (%): 307.1 [M+]+ (100).
1-(4-(1-Adamantyl)phenyl)-1H-imidazolium iodide (10) was
prepared from compound 9 (0.1829 g, 0.66 mmol) to yield a
colorless powder (0.1246 g, 0.30 mmol, yield 45%). Mp: 194−196
°C. Anal. Calcd for C20H215IN2: C, 57.15; H, 5.99; N, 6.66. Found: C,
57.07; H, 5.93; N; 6.82. H NMR (CDCl3, 500 MHz): δ 10.52 (s,
1H), 7.54−7.68 (m, 5H), 7.53 (m, 1H), 4.25 (s, 3H), 2.11 (um, 3H),
1.89 (m, 6H), 1.77 (m, 6H). 13C{1H} NMR (CDCl3, 126 MHz): δ
154.3, 136.3, 132.0, 127.4, 124.6, 121.9, 120.9, 43.1, 37.7, 36.7, 36.6,
28.9. IR (KBr): 6453 (sh), 3404 (sh), 3148 (w), 3095 (sh), 3072
(m), 3044 (sh), 3005 (sh), 2928 (sh), 2911 (sh), 2900 (vs), 2885
(sh), 2847 (vs), 2679 (vw), 2656 (vw), 1571 (m), 1551 (m), 1516
(sh), 1508 (w), 1499(w), 1412 (vw), 1384 (w), 1369 (vw), 1359
(vw), 1343 (w), 1317 (vw), 1310 (sh), 1272 (vw), 1251 (vw), 1235
(vw), 1220 (w), 1107 (vw), 1103 (vw), 1077 (w), 1069 (sh), 1033
(vw), 1024 (vw), 1017 (vw), 1013 (vw), 978 (vw), 975(sh), 956
(vw), 938 (vw), 850 (vw), 841 (w), 809 (m), 727 (vw), 657 (vw),
641 (vw), 627 (sh), 614 (w), 610 (w), 564 (w), 551 (w) cm−1. ESI-
MS (pos.) m/z (%): 293.2 [M+]+ (100).
1-(4-(1-Adamantyl)phenyl)-1H-imidazole (9) was prepared
using a modified literature procedure.48 A brown dispersion of 8
(0.9955 g, 4.40 mmol), MeOH (32 mL), and 40% glyoxal solution
(185 μL, 1.62 mmol) was stirred for 19 h at room temperature. Then
NH4Cl (0.4705 g, 8.80 mmol) and a 37% HCHO solution (0.704 mL,
8.87 mmol) were added, and the reaction mixture was refluxed for 1 h
using an oil bath. After the reaction mixture cooled to room
temperature, concd H3PO4 (0.616 mL, 85%) was slowly added and
then the mixture was refluxed for 19 h. The solvent was evaporated,
the remaining oil was poured on an ice−water mixture, and the pH
was adjusted with 40% KOH solution to 10. This mixture was
extracted with AcOEt (5 × 30 mL), the combined organic phases
were washed with brine, and the solution was dried with Na2SO4
before evaporating the solvent under the reduced pressure. The
afforded mixture was purified by column chromatography (silica gel,
AcOEt/MeOH, 8:1, v-v) and obtained a crude product (Rf = 0.47)
that was purified by column chromatography (silica gel, AcOEt) to
yield a brown powder (0.1913 g, 0.69 mmol, yield 42%). Mp: 116−
118 °C. Anal. Calcd for C19H22N2: C, 81.97; H, 7.97; N, 10.06.
1
Found: C, 81.72; H, 8.12; N, 9.93. H NMR (CDCl3, 500 MHz): δ
8.27 (s, 1H), 7.50 (m, 2H), 7.38 (m, 2H), 7.32 (s, 1H), 7.31 (s, 1H),
2.13 (um, 3H), 1.93 (m, 6H), 1.79 (m, 6H). 13C{1H} NMR (CDCl3,
126 MHz): δ 152.2, 135.0, 134.2, 127.4, 126.8, 121.7, 119.1, 43.3,
36.8, 36.4, 29.0. IR (KBr): 3122 (w), 3114 (sh), 3059 (vw), 3040
(vw), 2941 (sh), 2928 (s), 2918 (sh), 2910 (vs), 2898 (sh), 2852
(m), 2845 (sh), 1610 (w), 1584 (vw), 1526 (s), 1518 (sh), 1509 (sh),
1490 (m), 1449 (w), 1371 (vw), 1345 (w), 1306 (m), 1266 (vw),
1257 (m), 1255 (sh), 1245 (w), 1110 (w), 1102 (w), 1057 (m), 1034
(w), 964 (w), 904 (w), 847 (w), 827 (w), 806 (m), 734 (w), 726 (w),
721 (w), 658 (sh), 656 (m), 652 (sh), 623 (vw), 552 (w) cm−1. ESI-
MS (pos.) m/z (%): 279.1 [M + H+]+ (100), 301.1 [M + Na+]+ (15),
557.3 [2·M + H+]+ (23), 579.4 [2·M + Na+]+ (15).
General Procedure toward Imidazolium Salts 5a, 5b, and 10.
Iodomethane (5 equiv, 0.95−3.30 mmol) was added into a solution of
imidazole 4a, 4b, or 9 (1 equiv, 0.19−0.66 mmol) in dry toluene (5
mL) under an inert atmosphere. The reaction mixture was stirred at
room temperature and monitored using TLC until the starting
imidazole disappeared. Then dry Et2O (2 mL) was added to
precipitate the imidazolium salt. The solid material was washed five
times with dry Et2O using a centrifuge. Dispersion in Et2O was
transferred into a round-bottom flask and solvent was evaporated to
obtain pure imidazolium salt, which was dried in a vacuum to
constant weight prior to supramolecular studies.
(4-(1-Adamantyl)phenyl)trimethylammonium iodide (11) was
prepared following a modified literature procedure.5b A mixture of 8
(0.1030 g, 0.45 mmol), NaHCO3 (0.0840 g, 1.00 mmol), CH3I
(0.205 mL, 3.30 mmol) and MeOH (32 mL) was refluxed for 8 h
using an oil bath. Then, the second portion of MeI (0.205 mL, 3.30
mmol) was added and the reaction mixture was refluxed for additional
8 h. The mixture was cooled down, concentrated under reduced
pressure and washed with hot acetone (20 mL) to obtain a white
solid. The solid crude product was washed with several portions of
CHCl3 to separate 11 from inorganic salts, and a yellowish powder
was obtained after evaporation of the solvent (0.1069 g, 0.26 mmol,
yield 59%). Mp: 186−188 °C. Anal. Calcd for C19H28IN: C 57.43, H
1
7.10, N 3.53 (%). Found C 57.31, H 7.12, N 3.49 (%). H NMR
(DMSO-d6, 500 MHz): δ 7.89 (d, 2H, J = 9 Hz), 7.59 (d, 2H, J = 9
Hz), 3.60 (s, 9H), 2.07 (brs, 3H), 1.88 (um, 6H), 1.75 (um, 6H).
13C{1H} NMR (DMSO-d6, 126 MHz): δ 152.7, 144.8, 126.3, 120.0,
56.4, 42.2, 35.9, 35.8, 28.1. IR (KBr): 3446 (s), 3016 (w), 3005 (sh),
1-(4-(1-Adamantyl)benzyl)-1H-imidazolium iodide (5a) was
prepared from compound 4a (0.0839 g, 0.29 mmol) to yield a
4493
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
2927 (sh), 2900 (vs), 2887 (sh), 2847 (s), 2818 (sh), 2678 (vw),
2658 (vw), 1634 (w), 1506 (w), 1456 (w), 1343 (vw), 1318 (vw),
1237 (vw), 1186 (sh), 1182 (sh), 1155 (vw), 1127 (sh), 1123 (sh),
1106 (sh), 1102 (sh), 1068 (vw), 1036 (vw), 1011 (w), 974 (vw),
957 (w), 947 (w), 862 (vw), 850 (sh), 845 (w), 827 (vw), 808 (w),
577 (w) cm−1. ESI−MS (pos.) m/z (%): 270.1 [M+]+ (100).
1-(4-(1-Adamantyl)phenyl)pyridinium chloride (12) was pre-
pared using a modified literature procedure.49 Compound 8 (0.2000
g, 0.88 mmol) and N-(2,4-dinitrophenyl)pyridinium chloride (0.198
g, 0.97 mmol) were dissolved in EtOH (10 mL). The resulting
solution was refluxed for 26 h using an oil bath. Subsequently, the
reaction mixture was cooled to room temperature and EtOH was
evaporated under reduced pressure. The crude product was
crystallized from AcOEt to yield a dark orange powder (0.1813 g,
0.56 mmol, yield 63%). Mp: 196−200 °C. Anal. Calcd for
C21H24ClN·1.2H2O: C, 72.58; H, 7.66; N, 4.03. Found: C, 72.69;
H, 7.81; N, 3.88. HRMS (ESI/TOF) m/z: M+ Calcd for C21H24N
290.1909; Found 290.1902. 1H NMR (DMSO-d6, 400 MHz): δ 9.39
(um, 2H), 8.79 (m, 1H), 8.31 (um, 2H), 7.85 (um, 2H), 7.72 (m,
2H), 2.10 (um, 3H), 1.93 (m, 6H), 1.77 (um, 6H). 13C{1H} NMR
(DMSO-d6, 101 MHz): δ 154.1, 146.3, 144.8, 140.4, 128.1, 126.6,
124.3, 42.3, 36.1, 35.9, 28.2. IR (KBr): 3423 (w), 2931 (sh), 2904
(m), 2808 (sh), 1625 (vs), 1616 (sh), 1581 (vs), 1564 (vs), 1543
(sh), 1510 (vs), 1476 (w), 1443 (vs), 1413 (w), 1343 (sh), 1335 (vs),
1332 (vs), 1316 (sh), 1335 (vs), 1332 (vs), 1316 (sh), 1241 (vw),
12202 (sh), 1190 (sh), 1175 (vs), 1135 (m), 1102 (w), 1078 (vw),
1033 (w), 1012 (m), 974 (w), 963 (sh), 917 (vw), 887 (vw), 870
(vw), 845 (sh), 835 (m), 802 (w), 781 (w), 741 (vw), 701 (vw), 680
(w), 644 (vw), 568 (sh), 558 (sh), 539 (m), 520 (sh), 428 (vw), 414
(vw) cm−1. ESI-MS (pos.) m/z (%): 290.1 [M+]+ (100).
Authors
̌
Josef Tomecek − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic; orcid.org/0000-0001-9801-7393
Andrea Cablová − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic
Aneta Hromádková − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic
Radek Marek − CEITEC − Central European Institute of
Technology, National Centre for Biomolecular Research,
Faculty of Science, and Department of Chemistry, Faculty of
Science, Masaryk University, 625 00 Brno, Czech Republic;
orcid.org/0000-0002-3668-3523
Ivo Durník − CEITEC − Central European Institute of
Technology and National Centre for Biomolecular Research,
Faculty of Science, Masaryk University, 625 00 Brno, Czech
Republic
̌
̌
Zdenka Prucková − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic; orcid.org/0000-0002-8327-6429
Michal Rouchal − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic; orcid.org/0000-0002-0117-4040
Lenka Dastychová − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic; orcid.org/0000-0003-3416-329X
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.0c02917.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.joc.0c02917
■
sı
Notes
The authors declare no competing financial interest.
1H, 13C NMR spectra, X-ray data, ITC and MS data, and
computational details (PDF)
ACKNOWLEDGMENTS
The authors thank Professor Marek Necas from Masaryk
■
Structures in xyz format (ZIP)
̌
University for X-ray diffraction analyses of compounds 3b and
5b. This work has received funding from the Internal Funding
Agency of the Tomas Bata University in Zlín (Grant No. IGA/
FT/2020/001) and the Czech Science Foundation (Grant.
No. 18-05421S). Part of the research was also done within the
CEITEC 2020 (LQ1601) project with the financial contribu-
tion made by the Ministry of Education, Youths and Sports of
the Czech Republic (MEYS CR) with special support paid
from the National Programme for Sustainability II funds.
CIISB, Instruct-CZ Center of Instruct-ERIC EU consortium,
funded by MEYS CR infrastructure project LM2018127, is
gratefully acknowledged for the financial support of the
measurements at the Core Facilities Josef Dadok National
NMR Center and X-ray Diffraction and Bio-SAXS. The
computational resources were supported by MEYS CR from
the Large Infrastructures for Research, Experimental Develop-
ment and Innovations project “e-Infrastructure CZ-
LM2018140”.
Accession Codes
CCDC 1956556−1956557 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
́
Jan Novotny − CEITEC − Central European Institute of
Technology, National Centre for Biomolecular Research,
Faculty of Science, and Department of Chemistry, Faculty of
Science, Masaryk University, 625 00 Brno, Czech Republic;
orcid.org/0000-0002-1203-9549; Phone: +420-
549496440; Email: jan.novotny@ceitec.muni.cz
Petr Kulhánek − CEITEC − Central European Institute of
Technology and National Centre for Biomolecular Research,
Faculty of Science, Masaryk University, 625 00 Brno, Czech
Republic; orcid.org/0000-0002-4152-6514;
REFERENCES
■
(1) (a) Barrow, S. J.; Kasera, S.; Rowland, M. J.; Barrio, J.;
Scherman, O. A. Cucurbituril-Based Molecular Recognition. Chem.
Rev. 2015, 115, 12320−12406. (b) Rekharsky, M. V.; Inoue, Y.
Complexation Thermodynamics of Cyclodextrins. Chem. Rev. 1998,
98, 1875−1918.
(2) Assaf, K. I.; Nau, W. M. Cucurbiturils: from synthesis to high-
affinity binding and catalysis. Chem. Soc. Rev. 2015, 44, 394−418.
Phone: +420-549495459; Email: kulhanek@
chemi.muni.cz
Robert Vícha − Department of Chemistry, Faculty of
Technology, Tomas Bata University in Zlín, 760 01 Zlín,
Czech Republic; orcid.org/0000-0002-5229-9863;
Phone: +420-576031103; Email: rvicha@utb.cz
4494
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(3) Schibilla, F.; Voskuhl, J.; Fokina, N. A.; Dahl, J. E. P.; Schreiner,
P. R.; Ravoo, B. J. Host−Guest Complexes of Cyclodextrins and
Nanodiamonds as a Strong Non-Covalent Binding Motif for Self-
Assembled Nanomaterials. Chem. - Eur. J. 2017, 23, 16059−16065.
(4) (a) Moghaddam, S.; Yang, C.; Rekharsky, M.; Ko, Y. H.; Kim,
K.; Inoue, Y.; Gilson, M. K. New Ultrahigh Affinity Host−Guest
Complexes of Cucurbit[7]uril with Bicyclo[2.2.2]octane and
Adamantane Guests: Thermodynamic Analysis and Evaluation of
M2 Affinity Calculations. J. Am. Chem. Soc. 2011, 133, 3570−3581.
(b) Liu, S. M.; Ruspic, C.; Mukhopadhyay, P.; Chakrabarti, S.; Zavalij,
P. Y.; Isaacs, l. The Cucurbit[n]uril Family: Prime Components for
Cyclodextrin−Curcubit[6]uril-Wheeled [4]Rotaxanes with an Encap-
sulated Photosensitizer. Org. Lett. 2017, 19, 898−901.
(14) Connors, K. A.; Pendergast, D. D. Microscopic Binding
Constants in Cyclodextrin Systems: Complexation of a-Cyclodextrin
with Sym-1,4-disubstituted Benzenes. J. Am. Chem. Soc. 1984, 106,
7607−7614.
(15) (a) Tootoonchi, M. H.; Sharma, G.; Calles, J.; Prabhakar, R.;
Kaifer, A. E. Cooperative Self-Assembly of a Quaternary Complex
Formed by Two Cucurbit[7]uril Hosts, Cyclobis(paraquat-p-phenyl-
ene), and a “Designer” Guest. Angew. Chem., Int. Ed. 2016, 55,
11507−11511. (b) Ke, C.; Strutt, N. L.; Li, H.; Hou, X.; Hartlieb, K.
J.; McGonigal, P. R.; Ma, Z.; Iehl, J.; Stern, C. L.; Cheng, C.; Zhu, Z.;
Vermeulen, N. A.; Meade, T. J.; Botros, Y. Y.; Stoddart, J. F.
Pillar[5]arene as a Co-Factor in Templating Rotaxane Formation. J.
Am. Chem. Soc. 2013, 135, 17019−17030. (c) Rekharsky, M. V.;
Yamamura, H.; Kawai, M.; Osaka, I.; Arakawa, R.; Sato, A.; Ko, Y. H.;
Selvapalam, N.; Kim, K.; Inoue, Y. Sequential Formation of a Ternary
Complex among Dihexylammonium, Cucurbit[6]uril, and Cyclo-
dextrin with Positive Cooperativity. Org. Lett. 2006, 8, 815−818.
(16) (a) Ooya, T.; Inoue, D.; Choi, H. S.; Kobayashi, Y.; Loethen,
S.; Thompson, D. H.; Ko, Y. H.; Kim, K.; Yui, N. pH-Responsive
Movement of Cucurbit[7]uril in a Diblock Polypseudorotaxane
Containing Dimethyl α-Cyclodextrin and Cucurbit[7]uril. Org. Lett.
2006, 8, 3159−3162. (b) Yuan, L.; Wang, R.; Macartney, D. H.
Binding Modes of Cucurbit[6]uril and Cucurbit[7]uril with a
Tetracationic Bis(viologen) Guest. J. Org. Chem. 2007, 72, 4539−
4542. (c) Wyman, I. W.; Macartney, D. H. Host-Guest Complexes
and Pseudorotaxanes of Cucurbit[7]uril with Acetylcholinesterase
Inhibitors. J. Org. Chem. 2009, 74, 8031−8038. (d) Sun, H.-L.; Zhang,
H.-Y.; Dai, Z.; Han, X.; Liu, Y. Insights into the Difference Between
Rotaxane and Pseudorotaxane. Chem. - Asian J. 2017, 12, 265−270.
(e) Lin, R.-L.; Li, R.; Shi, H.; Zhang, K.; Meng, D.; Sun, W.-Q.; Chen,
K.; Liu, J.-X. Symmetrical-Tetramethyl-Cucurbit[6]uril-Driven Move-
ment of Cucurbit[7]uril Gives Rise to Heterowheel [4]-
Pseudorotaxanes. J. Org. Chem. 2020, 85, 3568−3575.
Self-Sorting Systems. J. Am. Chem. Soc. 2005, 127, 15959−15967.
̌
́
(5) (a) Cao, L. P.; Sekutor, M.; Zavalij, P. Y.; Mlinaric-Majerski, K.;
Glaser, R.; Isaacs, L. Cucurbit[7]uril·Guest Pair with an Attomolar
Dissociation Constant. Angew. Chem., Int. Ed. 2014, 53, 988−993.
̌
̌
(b) Sekutor, M.; Molcanov, K.; Cao, L. P.; Isaacs, L.; Glaser, R.;
́
Mlinaric-Majerski, K. Design, Synthesis, and X-ray Structural Analyses
of Diamantane Diammonium Salts: Guests for Cucurbit[n]uril
(CB[n]) Hosts. Eur. J. Org. Chem. 2014, 2014, 2533−2542.
(6) Jelínková, K.; Surmová, H.; Matelová, A.; Prucková, Z.; Rouchal,
̌
M.; Dastychová, L.; Necas, M.; Vícha, R. Cubane Arrives on the
Cucurbituril Scene. Org. Lett. 2017, 19, 2698−2701.
(7) Rekharsky, M. V.; Mori, T.; Yang, C.; Ko, Y. H.; Selvapalam, N.;
Kim, H.; Sobransingh, D.; Kaifer, A. E.; Liu, S.; Isaacs, L.; Chen, W.;
Moghaddam, S.; Gilson, M. K.; Kim, K.; Inoue, Y. A synthetic host−
guest system achieves avidin-biotin affinity by overcoming enthalpy−
entropy compensation. Proc. Natl. Acad. Sci. U. S. A. 2007, 104,
20737−20742.
́
(8) (a) Chyba, J.; Novák, M.; Munzarová, P.; Novotny, J.; Marek, R.
Through-Space Paramagnetic NMR Effects in Host-Guest Com-
plexes: Potential Ruthenium(III) Metallodrugs with Macrocyclic
Carriers. Inorg. Chem. 2018, 57, 8735−8747. (b) Malali, S.; Chyba, J.;
̌
́
Knor, M.; Horní, M.; Necas, M.; Novotny, J.; Marek, R. Zwitterionic
Ru(III) Complexes: Stability of Metal−Ligand Bond and Host−Guest
Binding with Cucurbit[7]uril. Inorg. Chem. 2020, 59, 10185−10196.
(17) Olah, G. A.; Farooq, O.; Farnia, S. M. F.; Wu, A. Aromatic
substitution. 58. Boron tris(triflate)-catalyzed adamantylation of
benzene and toluene with 1- and 2-haloadamantanes and
adamantanoyl chlorides. Isomerization of phenyl- and tolyladaman-
tanes. J. Org. Chem. 1990, 55, 1516−1522.
̌
̌
(9) Kolman, V.; Marek, R.; Strelcová, Z.; Kulhánek, P.; Necas, M.;
̌
̌
̌
Svec, J.; Sindelár, V. Electron Density Shift in Imidazolium
Derivatives upon Complexation with Cucurbit[6]uril. Chem. - Eur. J.
2009, 15, 6926−6931.
̌
(10) Branná, P.; Cernochová, J.; Rouchal, M.; Kulhánek, P.;
(18) Cheng, W. C.; Kurth, M. J. The Zincke Reaction. A Review.
Org. Prep. Proced. Int. 2002, 34, 585−608.
́
̌
̌
Babinsky, M.; Marek, R.; Necas, M.; Kuritka, I.; Vícha, R. Cooperative
Binding of Cucurbit[n]urils and β-Cyclodextrin to Heteroditopic
Imidazolium-Based Guests. J. Org. Chem. 2016, 81, 9595−9604.
(11) (a) Ding, C.; Liu, Y.; Wang, T.; Fu, J. Triple-stimuli-responsive
nanocontainers assembled by water-soluble pillar[5]arene-based
pseudorotaxanes for controlled release. J. Mater. Chem. B 2016, 4,
2819−2827. (b) Wang, M.-D.; Chen, T.; Ding, C.-D.; Fu, J.-J.
Mechanized silica nanoparticles based on reversible bistable [2]-
pseudorotaxanes as supramolecular nanovalves for multistage pH-
controlled release. Chem. Commun. 2014, 50, 5068−5071. (c) Sojka,
(19) Rössler, S. L.; Jelier, B. J.; Tripet, P. F.; Shemet, A.; Jeschke, G.;
Togni, A.; Carreira, E. E. Pyridyl Radical Cation for C−H Amination
of Arenes. Angew. Chem., Int. Ed. 2019, 58, 526−531.
(20) Hettiarachchi, D. S. N.; Macartney, D. H. Cucurbit [7] uril
host−guest complexes with cationic bis(4,5-dihydro-1H-imidazol-2-
yl) guests in aqueous solution. Can. J. Chem. 2006, 84, 905−914.
̌
̌
(21) Cernochová, J.; Branná, P.; Rouchal, M.; Kulhánek, P.; Kuritka,
I.; Vícha, R. Determination of Intrinsic Binding Modes by Mass
Spectrometry: Gas-Phase Behavior of Adamantylated Bisimidazolium
Guests Complexed to Cucurbiturils. Chem. - Eur. J. 2012, 18, 13633−
13637.
̊
̌
̌
M.; Fojtu, M.; Fialová, J.; Masarík, M.; Necas, M.; Marek, R. Locked
and Loaded: Ruthenium(II)-Capped Cucurbit[n]uril-Based Rotax-
anes with Antimetastatic Properties. Inorg. Chem. 2019, 58, 10861−
10870.
(22) Sahai, R.; Loper, G. L.; Lin, S. H.; Eyring, H. Investigation of
the Composition and Formation Constant of Molecular Complexes.
Proc. Natl. Acad. Sci. U. S. A. 1974, 71, 1499−1503.
(12) (a) Wang, B.; Han, J.; Bender, M.; Hahn, S.; Seehafer, K.; Bunz,
U. H. F. Poly(para-phenyleneethynylene)-Sensor Arrays Discriminate
22 Different Teas. ACS Sens. 2018, 3, 504−511. (b) Hu, J.-P.; He, J.-
X.; Fang, H.; Yang, H.-H.; Zhang, Q.; Lin, Q.; Yao, H.; Zhang, Y.-M.;
Wei, T.-B.; Qu, W.-J. A novel pillar[5]arene-based emission enhanced
supramolecular sensor for dual-channel selective detection and
separation of Hg2+. New J. Chem. 2020, 44, 13157−13162. (c) Liu,
Y.-C.; Nau, W. M.; Hennig, A. A supramolecular five-component relay
switch that exposes the mechanistic competition of dissociative versus
associative binding to cucurbiturils by ratiometric fluorescence
monitoring. Chem. Commun. 2019, 55, 14123−14126.
(23) Havemeyer, R. N. Freezing Point Curve of Dimethyl
SulfoxideWater Solutions. J. Pharm. Sci. 1966, 55, 851−853.
(24) Johnson, E. S. In Advances in Magnetic Resonance; Waugh, J. S.,
Ed.; Academic Press: New York, 1956; Vol. 1, pp 64−68.
(25) Jeener, J.; Meier, B. H.; Bachmann, P.; Ernst, R. R. Investigation
of exchange processes by two-dimensional NMR spectroscopy. J.
Chem. Phys. 1979, 71, 4546−4553.
̌
(26) (a) Kulkarni, S.; Jelínková, K.; Necas, M.; Prucková, Z.;
Rouchal, M.; Dastychová, L.; Kulhánek, P.; Vícha, R. A Photo-
chemical/Thermal Switch Based on 4,4′-Bis(benzimidazolio)stilbene:
Synthesis and Supramolecular Properties. ChemPhysChem 2020, 21,
(13) Yan, Z.; Huang, Q.; Liang, W.; Yu, X.; Zhou, D.; Wu, W.;
Schruma, J. J.; Yang, C. Enantiodifferentiation in the Photo-
isomerization of (Z,Z)-1,3-Cyclooctadiene in the Cavity of γ-
́
2084−2095. (b) Babjaková, E.; Branná, P.; Kuszynska, M.; Rouchal,
M.; Prucková, Z.; Dastychová, L.; Vícha, J.; Vícha, R. An adamantane-
4495
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
based disubstituted binding motif with picomolar dissociation
constants for cucurbit[n]urils in water and related quaternary
assemblies. RSC Adv. 2016, 6, 105146−105153. (c) Noujeim, N.;
Jouvelet, B.; Schmitzer, A. R. Formation of Inclusion Complexes
between 1,1′-Dialkyl-3,3′-(1,4-phenylene)bisimidazolium Dibromide
Salts and Cucurbit[7]uril. J. Phys. Chem. B 2009, 113, 16159−16168.
(d) Elie, C.-R.; Noujeim, N.; Pardin, C.; Schmitzer, A. R. Uncovering
new properties of imidazolium salts: Cl− transport and supramolecular
regulation of their transmembrane activity. Chem. Commun. 2011, 47,
1788−1790.
2015, 119, 1129−1151. (b) Darve, E.; Rodriguez-Gómez, D.;
Pohorille, A. Adaptive biasing force method for scalar and vector
free energy calculations. J. Chem. Phys. 2008, 128, 144120/1−13.
(41) (a) Raiteri, P.; Laio, A.; Gervasio, F. L.; Micheletti, C.;
Parrinelo, M. Efficient Reconstruction of Complex Free Energy
Landscapes by Multiple Walkers Metadynamics. J. Phys. Chem. B
̀
2006, 110, 3533−3539. (b) Minoukadeh, K.; Chipot, C.; Lelievre, T.
Potential of Mean Force Calculations: A Multiple-Walker Adaptive
Biasing Force Approach. J. Chem. Theory Comput. 2010, 6, 1008−
1017.
̌
̌
̌
(42) Kulhánek, P.; Stepán, J.; Fuxreiter, M.; Mones, L.; Strelcová, Z.;
(27) (a) Fokina, N. A.; Tkachenko, B. A.; Dahl, J. E. P.; Carlson, R.
M. K.; Fokin, A. A.; Schreiner, P. R. Synthesis of diamondoid
carboxylic acids. Synthesis 2012, 44, 259−264. (b) Fokina, N. A.;
Tkachenko, B. A.; Merz, A.; Serafin, M.; Dahl, J. E. P.; Carlson, R. M.
K.; Fokin, A. A.; Schreiner, P. R. Functionalized nanodiamonds, Part
8. Hydroxy derivatives of diamantane, triamantane, and [121]-
tetramantane: selective preparation of bis-apical derivatives. Eur. J.
Org. Chem. 2007, 2007, 4738−4745.
̌
Petrek, M. PMFLib − A Toolkit for Free Energy Calculations, https://
github.com/kulhanek/pmflib.
(43) Mones, L.; Bernstein, N.; Csányi, G. Exploration, Sampling,
And Reconstruction of Free Energy Surfaces with Gaussian Process
Regression. J. Chem. Theory Comput. 2016, 12, 5100−5110.
̈
(44) Stetter, H.; Schwarz, M.; Hirschhorn, A. Uber Verbindungen
mit Urotropin-Struktur, XII. Monofunktionelle Adamantan-Derivate.
Chem. Ber. 1959, 92, 1629−1635.
(28) Mosset, P.; Grée, R. Indium-Catalyzed Friedel−Crafts
Alkylation of Monosubstituted Benzenes by 1-Bromoadamantane.
Synlett 2013, 24, 1142−1146.
(45) (a) Matsubara, K.; Ishibashi, T.; Koga, Y. C−F Bond-Cleavage
Reactions of Fluoroalkanes with Magnesium Reagents and without
Metal Catalysts. Org. Lett. 2009, 11, 1765−1768. (b) Chalais, S.;
Cornélis, A.; Gerstmans, A.; Kołodziejski, W.; Laszlo, P.; Mathy, A.;
Métra, P. Direct Clay-Catalyzed Friedel−Crafts Arylation and
Chlorination of the Hydrocarbon Adamantane. Helv. Chim. Acta
1985, 68, 1196−1203.
(29) Rouchal, M.; Matelová, A.; Pires de Carvalho, F.; Bernat, R.;
̌
́
̌
́
Grbic, G.; Kuritka, I.; Babinsky, M.; Marek, R.; Cmelík, R.; Vícha, R.
Adamantane-bearing benzylamines and benzylamides: novel building
blocks for supramolecular systems with finely tuned binding
properties towards β-cyclodextrin. Supramol. Chem. 2013, 25, 349−
361.
(46) Stepanov, F. N.; Dikolenko, E. I.; Danilenko, G. I. Adamantane
and its derivatives. VI. Substitution reactions in aryladamantanes. Zh.
Obshch. Khim. 1966, 2, 640−643.
(47) Kogay, B. E.; Sokolenko, W. A. Reactions of 1.3-
dehydroadamantane (3.3.1-propellane system) with ch- and nh-
acids. Tetrahedron Lett. 1983, 24, 613−616.
(48) Stringer, B.; Quan, L.; Barnard, P.; Wilson, D.; Hogan, C.
Iridium Complexes of N-Heterocyclic Carbene Ligands: Investigation
into the Energetic Requirements for Efficient Electrogenerated
Chemiluminescence. Organometallics 2014, 33, 4860−4872.
(49) Song, Y.; Huang, X.; Hua, H.; Wang, Q. The synthesis of a rigid
conjugated viologen and its cucurbituril pseudorotaxanes. Dyes Pigm.
2017, 137, 229−235.
(30) Mori, S.; Takeuchi, Y.; Tanatani, A.; Kagechika, H.; Fujii, S.
Development of 1,3-diphenyladamantane derivatives as nonsteroidal
progesterone receptor antagonists. Bioorg. Med. Chem. 2015, 23, 803−
809.
̌
̌
(31) Vícha, R.; Kuritka, I.; Rouchal, M.; Jezková, V.; Zijerhut, A.
Directing effects in nitration of 1-adamantyl bearing aromatic ketones.
ARKIVOC 2009, 2009 (xii), 60−80.
̌
(32) Vícha, R.; Rouchal, M.; Kozubková, Z.; Kuritka, I.; Marek, R.;
Branná, P.; Cmelík, R. Novel Adamantane-Bearing Anilines and
̌
Properties of Their Supramolecular Complexes with β-Cyclodextrin.
Supramol. Chem. 2011, 23, 663−677.
(33) (a) Overhauser, A. W. Polarization of Nuclei in Metals. Phys.
Rev. 1953, 92, 411−415. (b) Kaiser, R. Use of the Nuclear
Overhauser Effect in the Analysis of High-Resolution Nuclear
Magnetic Resonance Spectra. J. Chem. Phys. 1963, 39, 2435−2442.
(c) Kaiser, R. Intermolecular Nuclear Overhauser Effect in Liquid
Solutions. J. Chem. Phys. 1965, 42, 1838−1839. (d) Anet, F. A. L.;
Bourn, A. J. R. Nuclear Magnetic Resonance Spectral Assignments
from Nuclear Overhauser Effects. J. Am. Chem. Soc. 1965, 87, 5250−
5251. (e) Neuhaus, D.; Williamson, M. The Nuclear Overhauser Effect
in Structural and Conformational Analysis; VCH Publisher, New York,
1989.
(34) Linden, A.; Mariz, R.; Blumentritt, S.; Dorta, R. CSD Commun.
2020, 2003919.
(35) (a) CrystalClear, The Woodlands, TX, 2014. (b) CrysAlisPro.
Rigaku Oxford Diffraction, Oxford, UK, 2015.
(36) (a) Sheldrick, G. M. Acta Crystallogr. 2015, A71, 3.
(b) Sheldrick, G. M. Acta Crystallogr. 2015, C71, 3.
(37) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.
A. Case: Development and testing of a general amber force field. J.
Comput. Chem. 2004, 25, 1157−1174.
(38) Kirschner, K. N.; Yongye, A. B.; Tschampel, S. M.; González-
Outeirino, J.; Daniels, C. R.; Foley, B. L.; Woods, R. J. GLYCAM06: a
̃
generalizable biomolecular force field. J. Comput. Chem. 2008, 29,
622−655.
(39) Case, D. A.; Betz, R. M.; Botello-Smith, W.; Cerutti, D. S.;
Cheatham, T. E., III; Darden, T. A.; Duke, R. E.; Giese, T. J.; Gohlke,
H.; Goetz, A. W.; et al.. AMBER 16; University of California: San
Francisco, 2016.
̀
(40) (a) Comer, J.; Gumbart, J. C.; Hénin, J.; Lelievre, T.; Pohorille,
A.; Chipot, C. The Adaptive Biasing Force Method: Everything You
Always Wanted To Know but Were Afraid To Ask. J. Phys. Chem. B
4496
https://dx.doi.org/10.1021/acs.joc.0c02917
J. Org. Chem. 2021, 86, 4483−4496