Enantioselective molecular recognition of chiral organic ammonium ions and amino acids using cavitand-salen-based receptors

Article abstract of DOI:10.1002/ejoc.201100955

Two new receptors, a cavitand-salen (2) and a uranyl-cavitand-salen (3), for the selective molecular recognition of chiral ammonium ion pairs, where the amino acid is the countercation or counteranion of the ion pair, were designed and synthesized. UV/Vis

Full text of DOI:10.1002/ejoc.201100955

FULL PAPER
DOI: 10.1002/ejoc.201100955
Enantioselective Molecular Recognition of Chiral Organic Ammonium Ions and
Amino Acids Using Cavitand–Salen-Based Receptors
Maria E. Amato,[a] Francesco P. Ballistreri,[a] Salvatore D’Agata,[a] Andrea Pappalardo,[a]
Gaetano A. Tomaselli,*[a] Rosa M. Toscano,[a] and Giuseppe Trusso Sfrazzetto[a]
Dedicated to Professor Gianfranco Scorrano on the occasion of his 72nd birthday
Keywords: Amino acids / Enantioselectivity / Chirality / Molecular recognition / Cavitands
Two new receptors, a cavitand–salen (2) and a uranyl–cavit-
and–salen (3), for the selective molecular recognition of chi-
ral ammonium ion pairs, where the amino acid is the
countercation or counteranion of the ion pair, were designed
and synthesized. UV/Vis measurements indicate the forma-
tion of 1:1 host–guest complexes with high association con-
stants and good to excellent enantiomeric discriminations.
NMR spectroscopic experiments confirmed cation–π interac-
tions between the organic cations and the π-electron-rich
cavity, leading to the stability of the complexes, as well as
the coordination of the amino acid carboxylate anion to the
uranyl metal center. Extraction experiments showed that a
racemic mixture of (R,S)-α-methylbenzyltrimethylammonium
iodides undergoes chiral resolution by 2 with high selectivity.
uranyl dication acting as a Lewis acidic coordination site,
whereas the organic cation may be bound by using π-rich
frameworks developing cation–π interactions.
Introduction
Many natural processes such as immunological respons-
es,[1a] the mechanism of drug action,[1b] and the storage and
retrieval of genetic information[1c] are regulated by enantio-
meric discrimination processes. The development of chiral
artificial receptors possessing chiral recognition properties
has attracted increasing attention because of their high sen-
sitivity and potential applications in pharmaceuticals,[2,3]
analysis,[4,5] biology,[6] catalysis,[7–9] and sensing.[10] Many
biological processes involve interactions between ammo-
nium ions and protein receptors, and a selective recognition
event controls the cellular response; for example, molecular
recognition of amino acids is of crucial importance for the
inhibition of biological processes. Therefore, there is great
interest in the design of receptors for ammonium ions and
amino acids. Important motifs for the recognition of am-
monium ions are hydrogen bonds and cation–π interactions,
and for this reason π-electron-rich macrocyclic cavities rep-
resent very suitable hosts.
In previous papers,[11,12] we have synthesized mononu-
clear and dinuclear uranyl chiral macrocyclic complexes, in-
corporating both a salen unit containing two phenyl rings
linked to a chiral diimine bridge and the (R)-BINOL unit,
which behave as efficient ditopic receptors for chiral quater-
nary ammonium salts. Likewise, a new heteroditopic chiral
uranyl–salen complex[13] incorporating two pyrenyl groups
was designed and synthesized in our laboratory for the re-
cognition of ammonium salts and tetrabutylammonium
and tetramethylammonium amino acids.
In this work we discuss a triquinoxaline-spanned cavit-
and decorated with a salen chiral framework as an enantio-
selective molecular receptor for chiral organic ammonium
salts and α-amino acid moieties. Cavitands are robust mole-
cules with a bowl shape consisting of four resorcinol rings
linked by four methylene bridges.[14] In order to host more
sizeable guests, the cavity wall size can be enlarged by inser-
tion of quinoxaline units, which make the cavity deeper and
at the same time more conformationally rigid and more li-
pophilic. The cavity can provide π-electron-rich regions able
to develop CH–π and π–π interactions, thus facilitating the
recognition of organic cations. The salen framework, be-
cause of the presence of two stereogenic carbon atoms in
the diimine bridge, generates a chiral pocket that, through
imine nitrogen and phenolic oxygen atoms, can coordinate
to the uranyl dication, creating, therefore, a Lewis acidic
site that can bind the counteranion of an ion pair.
Recently, we have devoted our attention to the synthesis
of new neutral heteroditopic receptors able to simulta-
neously bind chiral ammonium cations and their
counteranions. In our strategy, the anion is bound by a
[a] Dipartimento di Scienze Chimiche Università di Catania,
Viale A. Doria 6, 95125 Catania, Italy
Fax: +39-095-580138
E-mail: gtomaselli@unict.it
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100955.
5674
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5674–5680

Molecular Recognition of Organic Ammonium Ions and Amino Acids
Scheme 1. Synthesis of receptors 2 and 3. Reagents and conditions: (a) (1R,2R)-diphenylethylenediamine, 3,5-di-tert-butylsalicylaldehyde
hydrochloride,[15] Et3N, EtOH, r.t. (48%); (b) (AcO)2UO2, EtOH, r.t. (90%).
Results and Discussion
Cavitand–Salen Receptor 2
Cavitand–salen receptor 2 (Scheme 1) was synthesized by
treating aldehyde 1[15] with an iminoamino monochloride
obtained by reaction of (1R,2R)-diphenylethylenediamine
hydrochloride with 3,5-di-tert-butylsalicyladehyde.[16]
The structural characterization of 2 was carried out by
1
MS (ESI) measurements and by H and 13C as well as g-
COSY and T-ROESY NMR spectroscopy (see Supporting
Information). The disappearance of the aldehyde signal (δ
= 10.18 ppm) of 1 and the appearance of two new signals
Scheme 3. Amino acid ammonium carboxylates used as guests.
at δ = 8.38 and 8.57 ppm, due to the imine protons of the
formed Schiff base, support the presence of the salen frame-
work in the cavitand receptor. The methine CH protons (Ha
and Hb) of the salen bridge resonate as an AB system at δ
= 4.52 and 4.64 ppm, respectively (3J = 6.5 Hz). T-ROESY
spectra display ROE contacts between phenyl ring 1 of the
salen bridge and the quinoxaline wall (see Supporting In-
formation). The bridge methine CH protons of the cavitand
display chemical shift values in the range δ = 4–6 ppm and
are indicative of a vase conformation of the cavitand
itself.[14,17,18] All these data seem consistent with the struc-
ture reported in Scheme 1.
We have tested chiral tetramethylammonium iodides,
amino acid methyl esters with the amino group transformed
into a tetraalkylammonium iodide, and amino acid deriva-
tives as tetraalkylammonium carboxylates. Molecular re-
cognition studies were carried out by performing NMR
spectroscopy and UV/Vis measurements. We first investi-
gated the binding properties of receptor 2 towards both
(R)- and (S)-α-methylbenzyltrimethylammonium iodides
1
(MBntriMAI) by H NMR titrations in CDCl3 at 27 °C to
understand the role of the cation in the complexation pro-
cess. Since 2 does not have a preorganized binding site for
anions (even if the three OH groups as hydrogen-bond do-
nors might be involved in the recognition of the
counteranion) the counteranion will be driven mainly by an
ion pair effect.
Schemes 2 and 3 report the Q+X– salts employed as
guests in the molecular recognition process.
1H NMR titrations showed an upfield shift of the signals
of the trimethyl group and of the α methyl group for both
the (R) and (S) isomers, but larger changes were observed
in the chemical shift values of the (S) isomer. By way of
contrast, the aromatic protons of the (R) isomer did not
undergo any appreciable shift, whereas in the case of the
(S) isomer these protons underwent an upfield shift (Fig-
ure 1). We may conclude in both cases that the receptor
hosts the ammonium group inside the cavity but that the
(S) isomer is better accommodated. Molecular mechanics
calculations, using the MM2* force field of MacroModel
8.6,[19] appeared to be consistent with these findings (Fig-
ure 2).
Scheme 2. Ammonium and amino acid salts used as guests.
Eur. J. Org. Chem. 2011, 5674–5680
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5675

G. A. Tomaselli et al.
FULL PAPER
As Table 1 reports, the binding constant for the (S) en-
antiomer is higher than the corresponding value of the (R)
enantiomer (KS/KR = 15.9), pointing out that a very ef-
ficient chiral discrimination is working. In order to check
this finding, a solvent extraction experiment was carried
out. To a 1ϫ10–3 m aqueous solution of a racemic mixture
of (R)/(S)-MBntriMAI was added an equimolar solution of
1
receptor 2 in chloroform. H NMR spectroscopic measure-
ments show that in the H2O/CHCl3 biphasic system the two
salts are present only in water. The biphasic system contain-
ing the two salts and the receptor was stirred for 2 h at
22 °C and then, after separation, the specific rotation of the
aqueous phase was measured.
Figure 1. Selected regions of the 1H NMR spectra of (a) (R,S)-
MBntriMAI, (b) 2 + (R)-MBntriMAI (3 equiv.), (c) 2 + (S)-
MBntriMAI (3 equiv.).
Table 1. Binding constants (Ka) and Gibbs free-energy changes
(–ΔG0) for the complexation of (R)- and (S)-MBntriMAI with 2 in
CHCl3 at 27 °C.
Q+X–
Ka [m–1][a]
–ΔG0 [kJmol–1] KS/KR
(R)-MBntriMAI (3.18Ϯ0.27) ϫ104
25.6
32.5
15.9
(S)-MBntriMAI (5.07Ϯ0.34) ϫ105
[a] Binding constants for complexes of chiral ammonium salts with
host 2 calculated by Hyperquad 2006 (v. 3.1.60).
The outcome reported in Table 2 indicates that 2 is able
to extract the (S) enantiomer from the water almost quanti-
tatively (88%ee), which is in very good agreement with the
determined binding constant values. As the optimized
structures of Figure 2 show, the configuration of the
stereocenter plays an important role in establishing cation–
π as well as CH–π and π–π interactions; for the (R) isomer,
optimum orientation for interaction with the inner walls of
the cavitand cannot be adopted. In this case, for instance,
the phenyl ring can protrude through the open wall (salen
wall). In the case of the (S) isomer, the phenyl ring is posi-
tioned between the quinoxaline wall and phenyl ring 1 of
the salen wall, and probably the different interactions give
a substantial contribution to the observed chiral discrimi-
nation. However, a further contribution to chiral selectivity
is due also to the different interactions of the host–guest
during the approach of the guest to the receptor.[20] Fig-
ure 4 displays possible steric interactions between the
phenyl ring of the salen bridge (Ph1 in Scheme 1) and the
R group bonded to the stereogenic carbon atom of the
guest. We might envisage a larger unfavorable steric interac-
tion for the (R) isomer (R = phenyl) than for the (S) isomer
(R = methyl). Finally, the possible location of the iodide
anion near the OH region is counterbalanced by the con-
straint to maintain an optimal equilibrium distance of the
cation–anion of the ion pair, as this can affect differently
the binding energy of the two guests and could be responsi-
ble for further discrimination.
Figure 2. Optimized structures of the (left) 2 ʚ (S)-MBntriMAI
and (right) 2 ʚ (R)-MBntriMAI complexes.
Unfortunately, NMR titrations do not allow quantitative
determinations to be performed because of the complexity
and overlap of many signals as well as the increased relax-
ation time. Therefore, the binding constants were obtained
by UV/Vis titrations. The UV/Vis spectrum of 2 displays an
absorbance band at 330 nm (π–π* quinoxaline cavity transi-
tion) and at 450 nm (salen framework absorption), which
undergo modifications upon the addition of the guest.
Both (R)- and (S)-MBntriMAI yield complexes with a
1:1 stoichiometry but with different affinities. A representa-
tive example of a titration curve is shown in Figure 3 to-
gether with the corresponding Job plot.
In order to enlarge the scope of our receptor we tested
some amino acid derivatives as guests. The amino acids,
treated with CH3I (see Supporting Information), were con-
verted into the corresponding methyl esters, whereas their
amino groups were permethylated and converted into tetra-
alkylammonium iodides. Therefore, the driving force for the
Figure 3. UV/Vis titration of (S)-MBntriMAI with 2 in CHCl3 at
27 °C. The inset displays the corresponding Job plot.
5676
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5674–5680

Molecular Recognition of Organic Ammonium Ions and Amino Acids
Table 2. Polarimetric measurements at 22 °C of aqueous solutions
containing the pure (R)- and (S)-MBntriMAI salts, respectively,
and the salt left after extraction by 2.
Q+X–
[α]2D2
(R)-MBntriMAI
(S)-MBntriMAI
Aqueous solution (after extraction)
+54.7
–54.7
+48.4
Figure 5. Optimized structures of the (left) 2 ʚ d-Trp-I and (right)
2 ʚ l-Trp-I complexes (hydrogen atoms omitted for clarity).
framework is now the counteranion of the ion pair and
therefore the main cation–π interactions do not involve the
amino acid residue directly, because the organic anion now
must be “towed” towards the cavity by the cation through
an ion-pair effect. The binding constants reported in
Table 4 are nearly one order of magnitude lower than those
observed in the previous cases and appear in agreement
with the formation of weaker association complexes. The d-
and l-Trp-TBA pair (KD/KL = 0.80) displays nearly the
same affinity, presumably because the carboxylate
counteranion gives a scarce contribution to the selectivity.
Figure 4. Steric interactions in the recognition of (R)-MBntriAI
(hydrogen atoms and aliphatic chains of receptor omitted for clar-
ity).
molecular recognition of these guests is the cation–π inter-
action between the ammonium cation and the π-electron-
rich cavity of the receptor. Table 3 reports the pertinent af-
finity constant values obtained by UV/Vis titrations. A
strong chiral discrimination is working for the d- and l-
Trp-I pair (KD/KL = 14.7) and for the d- and l-Phe-I (KD/
KL = 0.11) pair, whereas the selectivity is lower for the d-
and l-Ala-I pair (KD/KL = 3.9). The high binding affinities
and the remarkable selectivity are comparable to those ob-
served for the ammonium salts reported in Table 1, proba-
bly because there is a good structural analogy between the
two series of guests.
Table 4. Binding constants (Ka) and Gibbs free-energy changes
(–ΔG0) for the complexation of amino acid tetrabutylammonium
(TBA) or tetramethylammonium (TMA) carboxylates with 2 in
CHCl3 at 27 °C.
Q+X–
Ka [m–1][a]
–ΔG0 [kJmol–1]
KD/KL
d-Phe-TBA
l-Phe-TBA
(5.09Ϯ0.75) ϫ104
26.8
23.1
26.1
27.1
25.8
26.3
26.5
23.7
3.5
(1.44Ϯ0.11) ϫ104
d-Phe-TMA (4.03Ϯ0.24) ϫ104
l-Phe-TMA (5.59Ϯ0.11) ϫ104
0.72
0.80
3.0
d-Trp-TBA
l-Trp-TBA
d-Ala-TBA
l-Ala-TBA
(3.31Ϯ0.17) ϫ104
(4.19Ϯ0.25) ϫ104
(4.35Ϯ0.34) ϫ104
(1.45Ϯ0.27) ϫ104
Table 3. Binding constants (Ka) and Gibbs free-energy changes
(–ΔG0) for the complexation of tetraalkylammonium iodide amino
acids methyl esters with 2 in CHCl3 at 27 °C.
[a] Binding constants for complexes of chiral ammonium salts with
host 2 calculated by Hyperquad 2006 (v. 3.1.60).
Q+X–
Ka [m–1][a]
–ΔG0 [kJmol–1]
KD/KL
d-Phe-I
l-Phe-I
d-Trp-I
l-Trp-I
d-Ala-I
l-Ala-I
(1.55Ϯ0.39) ϫ105
(1.59Ϯ0.32) ϫ106
(1.33Ϯ0.26) ϫ106
(9.07Ϯ0.62) ϫ104
(7.08Ϯ0.85) ϫ105
(1.79Ϯ0.13) ϫ105
29.6
35.5
34.9
28.2
33.3
29.9
0.11
1H NMR spectroscopic measurements seem to support
this interpretation because we observed an upfield shift of
the tetrabutyl cation signals (see Supporting Information)
and a downfield shift of the CH methine protons of the
cavitand, indicative of the hosting of the cation by the cav-
ity (Figure 6). On the other hand, the signals of the two Ha
and Hb protons of the chiral salen bridge (Scheme 1) un-
14.7
3.9
[a] Binding constants for complexes of chiral ammonium salts with
host 2 calculated by Hyperquad 2006 (v. 3.1.60).
In fact, both of them are ammonium cations with the dergo an upfield shift indicating that the carboxylate anion
nitrogen atom linked to a carbon stereocenter bearing an is located near the bridge of the salen wall (probably in
aromatic residue or an α-carbon atom with an aromatic res- proximity to the OH groups, which, due to their hydrogen-
idue, and in both cases the driving force for the molecular donor abilities, might provide a suitable location for the
recognition is the affinity of the cation for the π-electron- carboxylate anion) and therefore quite outside the cavity.
rich cavity. Figure 5 displays the optimized structures for d- Some selectivity is observed with the d- and l-Phe-TBA
and l-Trp-I.
(KD/KL = 3.5) pair and with the d- and l-Ala-TBA (KD/KL
Further information on the molecular recognition of re- = 3.0) pair. Different stabilization by hydrogen bonding of
ceptor 2 toward amino acids was collected by converting the counteranion (which has two hydrogen-bond acceptor
the amino acids into their tetraalkylammonium carboxylate sites, that is, the carboxylate and the amino groups) due to
salts and measuring the pertinent binding constants. In this the different configurations of the stereocenter might be
case, we expected lower selectivity because the amino acid also responsible of the selectivity.[21]
Eur. J. Org. Chem. 2011, 5674–5680
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5677

G. A. Tomaselli et al.
FULL PAPER
Figure 6. Selected region of the 1H NMR spectra of (top) receptor
2 and (bottom) 2 + d-Ala-TBA (3 equiv.) showing the upfield shift
of the Ha and Hb protons of the salen bridge and the downfield
shift of the signals (black diamond) relative to the methine protons
of the cavitand.
1
Figure 7. Selected region of the H NMR spectra of (a) 3 and (b)
3 + d-Ala-TBA (3 equiv.).
Heteroditopic Uranyl–Cavitand–Salen Receptor 3
By complexation with uranyl acetate, receptor 2 was uti-
lized to prepare uranyl–cavitand–salen 3 (Scheme 1), able
to work as a heteroditopic receptor because of the presence
of both a Lewis acid uranyl center, which can bind the
counteranion of an ion pair, and the electron-rich cavity
able to recognize the corresponding countercation. Com-
plex 3 was characterized by MS (ESI) measurements and
by 1H NMR spectroscopy. The coordination of UO2 causes
a downfield shift of CH imine proton signals (from δ =
8.38/8.57 to δ = 9.02/9.34 ppm) as well as of the Ha and Hb
proton signals of the salen bridge (from δ = 4.52/4.64 to δ =
5.11/5.79 ppm), whose coupling constant 3J = 8.0 Hz would
indicate that the two phenyl rings of the salen bridge form
a dihedral angle of about 130°.
Figure 8. Optimized structures of the (left) 3 ʚ d-Phe-TBA and
(right) 3 ʚ l-Phe-TBA complexes (hydrogen atoms omitted for
clarity).
Table 5. Binding constants (Ka) and Gibbs free-energy changes
(–ΔG0) for the complexation of amino acid tetrabutylammonium
(TBA) or tetramethylammonium (TMA) carboxylates with 3 in
CHCl3 at 27 °C.
We have already reported[13] that amino acid carboxylate
ammonium salts can be hosted by UO2–salen heteroditopic
receptors, because the carboxylate anion is able to bind the
fifth equatorial coordination site of the uranyl(VI) ion.[22]
The coordination of the uranyl center by the carboxylate
anion is observed also with receptor 3, as NMR spectro-
scopic experiments show. In fact, the observed upfield shift
of the imine proton signals and the aromatic proton signal
in the ortho position to the free OH group of 3 upon NMR
titration seems to provide good evidence that the carboxyl-
ate anion is able to coordinate the uranyl metal center (Fig-
ure 7). At the same time, the α-CH2 signals of the tetra-
butylammonium countercation undergo an upfield shift in-
dicating that it can be accommodated inside the cavity of
the receptor (see Supporting Information).
Q+X–
Ka [m–1][a]
–ΔG0 [kJmol–1]
KD/KL
d-Phe-TBA
l-Phe-TBA
d-Phe-TMA
l-Phe-TMA
d-Trp-TBA
l-Trp-TBA
d-Ala-TBA
l-Ala-TBA
(7.15Ϯ0.94) ϫ105
(1.09Ϯ0.15) ϫ106
(2.10Ϯ0.25) ϫ106
(8.85Ϯ0.15) ϫ104
(4.69Ϯ0.51) ϫ104
(4.62Ϯ0.67) ϫ104
(2.05Ϯ0.14) ϫ105
(3.21Ϯ0.31) ϫ104
33.4
34.4
36.1
28.2
26.7
26.6
30.3
25.7
0.66
23.7
1.0
6.4
[a] Binding constants for complexes of chiral ammonium salts with
host 3 calculated by Hyperquad 2006 (v. 3.1.60).
The presence of a preorganized binding site in 3, which
can also accommodate the anion, increases the binding af-
Figure 8 reports the optimized structures of the ternary finity by nearly one order of magnitude. However, the ob-
complexes of 3 with d- and l-Phe-TBA, which appear to served chiral discrimination is lower and comparable to that
be consistent with NMR spectroscopic indications, showing determined with 2, because also in this case the amino acid
the binding of the cation and the anion inside the cavity is the counteranion, and therefore, it is accommodated in
and at the uranyl metal center.
the coordination sphere of the uranyl moiety and not inside
The binding constants, determined by UV/Vis measure- the cavity. Both the d- and l-Trp-TBA (KD/KL = 1.0) and
ments, are reported in Table 5 and are more or less compar- the d- and l-Phe-TBA (KD/KL = 0.66) pairs display nearly
able in magnitude to the values obtained for the permethyl- the same selectivity, and only the d- and l-Ala-TBA (KD/
ated amino acids but larger than those measured for the KL = 6.4) pair shows selective recognition. With the smaller
same ammonium carboxylates with receptor 2 that does not amino acid guest, coordination to the metal center appears
possess the uranyl center as the binding site for the carb- more susceptible to the different configurations of the car-
oxylate anion.
bon atom stereocenter and then the molecular recognition
5678
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5674–5680

Molecular Recognition of Organic Ammonium Ions and Amino Acids
is more selective. The possibility for the anion to bind the in our laboratory to improve our knowledge on the rules
fifth equatorial coordination site of the metal and at the governing the interactions of these cavitand–salen receptors
same time to be hydrogen bonded through the amino group with amino acid guests in order to design responsive host–
to the free OH of the receptor in more or less straightfor- guest systems, because of their potential applications in sep-
ward and might be responsible, at least partially, for the aration and in drug design.
selectivity. Figure 8 shows that only l-Phe-TBA can be hy-
drogen bonded to the OH group with the NH2 group.
A very strong cation effect is observed with the d- and
l-Phe-TMA pair. The replacement of TBA with the smaller
Experimental Section
TMA changes the selectivity dramatically from KD/KL
=
General Experimental Methods: NMR experiments were carried
out at 27 °C with a Varian UNITY Inova 500 MHz spectrometer
(1H at 499.88 MHz, 13C NMR at 125.7 MHz) equipped with pulse
field gradient module (Z axis) and a tunable 5 mm Varian inverse
detection probe (ID-PFG). Unless otherwise stated, NMR spectra
were obtained in CDCl3 solutions. The chemical shifts were refer-
enced by using the residual solvent signal as the internal standard.
1H NMR peak assignments follow from 2D-COSY experiments.
Mass spectra (ESI) were acquired with an ESI-MS Thermo-Finni-
gan LCQ-DECA using MeOH (positive ion mode). A JASCO V-
560 UV/Vis spectrophotometer equipped with a 1 cm path-length
cell was used for the measurements of Job plots and for the UV/
Vis titrations. d-/l-Phenylalanine, d-/l-tryptophan, d-/l-alanine,
and (1R,2R)-(+)-1,2-diphenylethylenediamine were commercially
available reagent-grade materials (Aldrich). Monoiminoamino-
(1R,2R)-diphenyl-3,5-di-tert-butylsalicylaldehyde and (R)-/(S)-
MBnTMAI were obtained as previously reported.[16] All chemicals
were reagent grade and were used without further purification.
0.66 to KD/KL = 23.7. This effect was already observed with
the pyrenyl receptor.[13]
At the moment we do not have a rational explanation
for this behavior, but the change from a loose to a tight ion
pair and the modification of the conformation of the recep-
tor upon decreasing the size of the guest cation probably
have some effect on the coordination and hydrogen bonding
of the anion, making the energy binding more sensitive to
the configuration of the stereocenter of the guest.
The observation that in most cases the host–guest ex-
change of the ammonium ion is fast on the NMR timescale,
whereas the receptor seems to be in slow exchange with the
complex (cation protons appear as a time-averaged signal,
but the imine protons of the free and complexed receptor
appear as separate signals) deserves some comment. Man-
dolini[23] observed a similar situation in the ion pair re-
cognition of quaternary ammonium ions by uranyl–salo-
Receptor 2: In a round bottom flask, to a solution of monoformyl
phen receptors. The rationale offered is based on the as- cavitand 1[15] (0.183 mmol) in absolute EtOH (20 mL) was added
monoimineamine-(1R,2R)-diphenyl-3,5-di-tert-butylsalicylalde-
hyde[16] (0.183 mmol) and triethylamine (0.366 mmol). The reaction
was stirred for 48 h at room temperature and monitored by TLC
(hexane/EtOAc, 80:20). The reaction was quenched by evaporation
of the solvent under reduced pressure, and receptor 2 was purified
by flash chromatography (hexane/EtOAc, 80:20). Yield: 48%. 1H
NMR (500 MHz, CDCl3): δ = 15.79 (s, 1 H, OH), 13.05 (s, 1 H,
OH), 9.55 (s, 1 H, OH), 8.56 (s, 1 H, CH=N), 8.38 (s, 1 H, Ar),
8.15 (s, 1 H, CH=N), 8.00 (d, J = 8.5 Hz, 1 H, Ar), 7.95 (d, J =
8.5 Hz, 1 H, Ar), 7.87 (d, J = 8.5 Hz, 1 H Ar), 7.80–7.83 (m, 2 H,
sumption that the recognition mechanism involves a slow
equilibrium between the receptor and the ion pair, which
equilibrates the receptor between its free and complexed
form, and a second fast equilibrium between the ternary
complex formed and a second external ion pair (ion-quartet
mechanism), which implies a fast cation exchange between
the ion pair complexed and the external ion pair. Also, our
data seem to support this mechanism and the fast cation
exchange observed with receptors 2 and 3 is favored by the
presence of a mobile fourth wall (salen wall) in the cavitand. Ar), 7.75 (d, J = 8.5 Hz, 1 H Ar), 7.71 (s, 1 H, Ar), 7.50–7.63 (m,
6 H, Ar), 7.48 (s, 1 H, Ar), 7.36 (s, 1 H, Ar), 7.21–7.24 (m, 3 H,
ArH), 7.18 (s, 2 H, Ar), 7.11–7.15 (m, 2 H, Ar), 7.03–7.09 (m, 3
H, Ar), 6.94–6.98 (m, 4 H, Ar), 6.66 (s, 1 H, Ar), 5.68 (t, J =
8.0 Hz, 1 H, CH methine), 5.63 (t, J = 8.0 Hz, 1 H, CH methine),
Conclusions
4.62–4.71 [m, 2 H, CH2(CH2)3CH3 and 1 H, CH methine], 4.52 (d,
J = 6.5 Hz, 1 H, CH chiral bridge), 4.46 (t, J = 8.0 Hz, 1 H, CH
methine), 4.00 (t, J = 8.0 Hz, 1 H, CH methine), 2.00–2.40 [m, 6
H, CH2(CH2)3CH3 and 2 H, CH2(CH2)3CH3], 1.36–1.57 [m, 22 H,
We have designed and synthesized cavitand–salen (2) and
cavitand–salen–uranyl (3) receptors for the molecular re-
cognition of chiral ammonium ion pairs, where the amino
acid is the countercation or the counteranion of the ion
pair. UV/Vis measurements and Job plots indicate the for-
CH2(CH2)3CH3], 1.27 (s, 9 H, tBu), 1.22–1.31 (m, 12 H), 0.92 [t, J
= 7.0 Hz, 3 H, CH2(CH2)3CH3], 0.89 [t, J = 7.0 Hz, 3 H, CH2(CH2)3-
mation of 1:1 host–guest complexes. Larger selectivity val- CH3], 0.74 [t, J = 7.0 Hz, 3 H, CH2(CH2)3CH3] ppm. 13C NMR
(125 MHz, CDCl3): δ = 168.0, 165.1, 162.2, 157.9, 157.7, 154.6,
153.6, 152.6, 152.4, 152.2, 152.1, 151.9, 148.8, 147.5, 147.0, 139.95,
139.91, 139.7, 139.6, 138.4, 138.3, 137.39, 137.31, 136.3, 136.2,
135.2, 134.0, 130.1, 129.6, 129.1, 128.8, 128.6, 128.3, 128.1, 128.0,
127.9, 127.8, 127.7, 127.6, 127.4, 127.2, 126.9, 126.7, 126.4, 126.1,
124.4, 122.9, 122.8, 118.6, 118.0, 117.8, 117.3, 116.3, 111.5, 106.8,
78.4, 76.2, 63.6, 34.9, 34.8, 34.6, 34.2, 34.1, 33.6, 33.4, 33.3, 33.2,
32.8, 32.3, 31.9, 31.6, 31.99, 31.38, 31.0, 29.5, 29.4, 29.3, 29.2, 29.0,
28.0, 27.9, 27.6, 22.7, 22.6, 22.4, 14.07, 14.06, 13.8 ppm. MS (ESI):
m/z = 1686 [M + H + C2H5OH]+. C106H112N8 (1498.10): calcd. C
ues are observed for ion pairs whose quaternary ammonium
cation is an amino acid, due to different π–π and CH–π
interactions, whereas the selectivity is substantially lower
when an amino acid carboxylate is the anion of the ion pair.
When the guest is an amino acid carboxylate ammonium
salt, the presence of a preorganized binding site in the re-
ceptor, which can also accommodate the anion, increases
the affinity but not the selectivity, even if a very high value
of selectivity is observed for the d- and l-Phe pair when
changing the cation from TBA to TMA. Work is in progress 77.53, H 6.87, N 6.82; found C 77.47, H 6.82, N 6.77.
Eur. J. Org. Chem. 2011, 5674–5680
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5679

G. A. Tomaselli et al.
FULL PAPER
[9] E. Manoury, P. Kalck, M. Urrutigoity, J. Mol. Catal. A 2007,
278, 102–106.
[10] a) Y. Kubo, S. Maeda, S. Tokita, M. Kuba, Nature 1996, 382,
522–524; b) Y. Shoji, K. Tashiro, T. Aida, J. Am. Chem. Soc.
2006, 128, 10690–10691; c) H. Guo, J. Kim, S. Kim, S. Chang,
W. Kim, Langmuir 2009, 25, 648–652.
Uranyl Receptor 3: To a solution of 2 (0.089 mmol) dissolved in
absolute ethanol (10 mL) was added uranyl acetate (0.128 mmol).
The reaction was stirred overnight at room temperature, and the
resulting solid was filtered and dried to yield 3 as a red powder
(90%). 1H NMR (500 MHz, CDCl3): δ = 9.34 (s, 1 H, CH=N),
9.02 (s, 1 H, CH=N), 8.39 (s, 1 H, Ar), 8.00 (d, J = 8.0 Hz, 1 H,
Ar), 7.92 (d, J = 8.0 Hz, 1 H, Ar), 7.82 (d, J = 8.0 Hz, 2 H, Ar),
7.75 (s, 1 H, Ar), 7.72 (s, 1 H, Ar), 7.67 (d, J = 8.0 Hz, 1 H, Ar),
7.51–7.63 (m, 7 H, Ar), 7.48 (s, 1 H, Ar), 7.42 (s, 1 H, Ar), 7.39 (s,
1 H, Ar), 7.33 (s, 1 H, Ar), 7.32 (s, 1 H, Ar), 7.16–7.21 (m, 7 H,
Ar), 7.09 (s, 1 H, Ar), 7.08 (s, 1 H, Ar), 7.04 (m, 3 H, Ar), 6.80 (s,
1 H, OH), 5.79 (d, J = 8.0 Hz, 1 H, CH chiral bridge), 5.65–5.70
[m, 3 H, CH methine and CH2(CH2)3CH3], 5.11 (d, J = 8.0 Hz, 1
H, CH chiral bridge), 4.64 (t, J = 7.5 Hz, 2 H, CH methine), 4.03
(t, J = 7.5 Hz, 1 H, CH methine), 2.23–2.45 [m, 8 H, CH2(CH2)3-
CH3], 1.68 (s, 9 H, CH3 tBu), 1.33–1.63 [m, 24 H, CH2(CH2)3CH3],
1.27 (s, 9 H, CH3 tBu), 0.94 [t, J = 6.5 Hz, 6 H, CH2(CH2)3CH3],
0.79 [t, J = 6.5 Hz, 3 H, CH2(CH2)3CH3], 0.65 [t, J = 6.5 Hz, 3 H,
CH2(CH2)3CH3] ppm. MS (ESI): m/z = 1955.4 [M + H +
[11] M. E. Amato, F. P. Ballistreri, A. Pappalardo, D. Sciotto, G. A.
Tomaselli, R. M. Toscano, Tetrahedron 2007, 63, 9751–9757.
[12] M. E. Amato, F. P. Ballistreri, S. Gentile, A. Pappalardo, G. A.
Tomaselli, R. M. Toscano, J. Org. Chem. 2010, 75, 1437–1443.
[13] F. P. Ballistreri, A. Pappalardo, G. A. Tomaselli, R. M. Tos-
cano, G. Trusso Sfrazzetto, Eur. J. Org. Chem. 2010, 3806–
3810.
[14] a) D. J. Cram, Science 1983, 219, 1177–1183; b) P. P. Castro, G.
Zhao, G. A. Masankay, C. Hernandez, L. M. Gutierrez-Tun-
stad, Org. Lett. 2004, 6, 333–336; c) S. Richeter, J. Rebek Jr.,
J. Am. Chem. Soc. 2004, 126, 16280–16281; d) P. Byron, J. Re-
bek Jr., Chem. Commun. 2005, 722–724; e) T. Iwasawa, E.
Mann, J. Rebek Jr., J. Am. Chem. Soc. 2006, 128, 9308–9309;
f) D. Ajami, J. Rebek Jr., Angew. Chem. 2008, 120, 6148; An-
gew. Chem. Int. Ed. 2008, 47, 6059–6061; g) J. Hou, D. Ajanni,
J. Rebek Jr., J. Am. Chem. Soc. 2008, 130, 7810–7811.
[15] F. P. Ballistreri, A. Pappalardo, G. A. Tomaselli, E. Vittorino,
S. Sortino, G. Trusso Sfrazzetto, New J. Chem. 2010, 34, 2828–
2834.
2H5OH]+, 1978.8 [M
+ Na +
C2H5OH]+. C106H110N8O11U
(1910.11): calcd. C 66.65, H 5.80, N 5.87; found C 66.61, H 5.77,
N 5.81.
Supporting Information (see footnote on the first page of this arti-
cle): 1H, 13C, g-COSY, T-ROESY, and 2D-DOSY NMR spectra,
MS (ESI), UV/Vis titration curves, and Job Plots.
[16] M. E. Amato, F. P. Ballistreri, A. Pappalardo, G. A. Tomaselli,
R. M. Toscano, J. D. Williams, Eur. J. Org. Chem. 2005, 3562–
3570.
[17] P. Skinner, F. Diederich, Helv. Chim. Acta 2001, 84, 2146–2153.
[18] Y. Yamakoshi, R. Schlittler, F. Diederich, J. Mater. Chem.
2001, 11, 2895–2897.
Acknowledgments
[19]
F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J.
Comput. Chem. 1990, 11, 440–467.
We thank the University of Catania for financial support
[20]
Rebek reported that four-wall cavitands are not suitable recep-
tors for ammonium cations bearing bulky aliphatic groups but
that an open side is an essential requisite so that they can work
(three wall receptors) (A. Lledò, R. J. Hooley, J. Rebek Jr., Org.
Lett. 2008, 10, 3669–3671). In our receptor, the open side is
the salen wall, and therefore, cations must cross it to go inside
the cavity; this determines the interactions governed by the
host–guest stereocenters, which contribute to chiral discrimi-
nation
The optimized structure for complex 2 ʚ d-Phe-TBA (see Sup-
porting Information) appears to be consistent with the NMR
spectroscopic data and seems to support these observations.
A pentagonal bipyramidal coordination geometry for the uran-
yl(VI) ion with two axial oxido groups and with the fifth equa-
torial site available for complexation with anionic monodentate
ligands X– has been previously indicated in these uranyl chiral
macrocyclic complexes.[11–13]
[1] a) J. M. Hopkin, Science 1997, 276, 18–19; b) R. Williams, L.
Chang, A. W. Mudge, A. J. Harwood, Nature 2002, 417, 292–
295; c) T. Hermann, D. J. Patel, Science 2000, 287, 820–825.
[2] a) S. H. Choi, K. M. Huh, T. Ooaya, N. Yui, J. Am. Chem.
Soc. 2003, 125, 6350–6531; b) A. Guerrero-Martinez, T. Mon-
toro, M. H. Vinas, G. Tardajos, J. Pharm. Sci. 2008, 97, 1484–
1498; c) B. He, Y. Shi, Am. Pharm. Rev. 2008, 11, 47–52.
[3] Y. Bae, S. Fukushima, A. Harada, K. Kataoka, Angew. Chem.
2003, 115, 4788; Angew. Chem. Int. Ed. 2003, 42, 4640–4643.
[4] a) W. H. Pirkle, T. C. Pochapsky, Chem. Rev. 1989, 89, 347–
362; b) W. Tang, S. Ng, Nat. Protoc. 2008, 3, 691–697.
[5] P. Mikus, K. Marakova, J. Marak, D. Kaniansky, I. Valaskova,
E. Havranek, J. Chromatogr. A 2008, 1179, 9–16.
[6] a) D. N. Reinhoudt (Ed.), Comprehensive Supramolecular
Chemistry, Pergamon, New York, 1996, vol. 10; b) M. Sper-
anza, F. Rondino, M. Satta, A. Paladini, A. Giardini, D. Ca-
tone, S. Piccirillo, Chirality 2009, 21, 119–144.
[7] A. Vargas, G. Santarossa, M. Iannuzzi, A. Baiker, J. Phys.
Chem. C 2008, 112, 10200–10208.
[21]
[22]
[23]
M. Cametti, M. Nissinen, A. Dalla Cort, L. Mandolini, K.
Rissanen, J. Am. Chem. Soc. 2007, 129, 3641–3648.
Received: June 30, 2011
[8] H. Tsukube, S. Shinoda, H. Tamiaki, Coord. Chem. Rev. 2002,
226, 227–234.
Published Online: August 31, 2011
5680
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5674–5680