Novel arificial receptors for alkylammonium ions with remarkable selectivity and affinity

Article abstract of DOI:10.1002/1521-3765(20000915)6:18<3399::AID-CHEM3399>3.0.CO;2-M

The benzene-based tripodal tris(oxazolines) have been developed as the most selective and strong receptors toward linear alkylammonium ions reported to date. Among six tris(oxazolines) based on 2,4,6-trimethylbenzene framework, the phenylglycinol-derived

Full text of DOI:10.1002/1521-3765(20000915)6:18<3399::AID-CHEM3399>3.0.CO;2-M

FULL PAPER
Novel Artificial Receptors for Alkylammonium Ions with Remarkable
Selectivity and Affinity
Sung-Gon Kim and Kyo Han Ahn*[a]
Abstract: The benzene-based tripodal
tris(oxazolines) have been developed as
the most selective and strong receptors
toward linear alkylammonium ions re-
ported to date. Among six tris(oxazo-
lines) based on 2,4,6-trimethylbenzene
framework, the phenylglycinol-derived
0.02). Both receptors 4 and 6 extract b-
more enhanced to 4000 with tris(oxazo-
line) 9. The enhanced binding affinity
and high selectivity observed with re-
ceptors 4 and related derivatives 7 ± 9
compared with others can be explained
by an optimized steric and electronic
environment provided by the phenyl
substituents, which has been unambigu-
ously demonstrated by X-ray crystallo-
graphic and 1H NMR spectroscopic
studies on the host ± guest complexes.
The new receptor system has several
unique features such as ready availabil-
ity, structural simplicity, and in particu-
lar, versatility in derivatization. By vir-
tue of these advantages, it can be readily
tailored as selective receptors toward
biologically important amines.
phenethylammonium ion from water
into chloroform almost completely.
When the benzene frame is changed
from 2,4,6-trimethylbenzene to 2,4,6-
triethylbenzene, dramatic changes in
the affinity as well as in the selectivity
are observed. The association constant
observed by tris(oxazoline) 8 toward
receptor 4 exhibits the largest associa-
‡
tion
constant
toward
nBuNH3
nBuNH3 approaches 108 mÀ1 and the
‡
(logKass ˆ 6.65 Æ 0.02), while a similar
‡
‡
‡
value toward tBuNH3
(logKass
ˆ
selectivity ratio of nBuNH3 /tBuNH3 is
3.80 Æ 0.01) compared with others,
increased to 2700. This selectivity is even
which corresponds to the selectivity
‡
‡
ratio of nBuNH3 /tBuNH3 as high as
ꢀ700. The tris(oxazoline) 6 that has
bare oxazoline ring exhibits still a large
Keywords: alkylamines
´ tripodal
oxazolines hydrogen bonds
´
´
host ± guest chemistry
recognition
´
molecular
association constant toward sterically
‡
hindered tBuNH3
(logKass ˆ 5.26 Æ
An efficient recognition system for alkylammonium ions
Introduction
such as n-butylammonium and b-phenethylammonium ions is
valuable in clinical application,[4] because they are the
structural motifs in biologically important amines such as
GABA and dopamine.[5] The selective recognition of isomeric
butylammonium ions has been reported recently by Chang[6]
and Pappalardo[7] using calixarene-based receptors. A dra-
matic increase in selectivity was observed in going from
calix[6]arene- to calix[5]arene-based receptors, demonstrat-
ing a crucial complementary interaction between a host and a
guest for a successful molecular recognition. In this regard,
our oxazoline receptors provide a great opportunity to
examine the steric and electronic interactions between the
receptor and guest molecules.
Recently, we have developed a benzene-based tripodal
‡
oxazoline receptor system for selective recognition of NH4
‡ [1]
over K , which has potential applicability in clinical and
environmental chemistry. One of the receptors exhibited a
‡
‡
large binding affinity toward NH4 and a remarkable NH4 -
‡ [2]
‡
selectivity over K . The NH4 -binding affinity was depend-
ent on the oxazoline substituents, which suggested that the
selective recognition of a certain type of alkylammonium ions
over others could be achieved by optimizing the steric and
electronic influence of the substituents. In designing synthetic
receptors for organic guests, the challenging problem is how to
rationally incorporate interesting guest structures into recep-
tor structures.[3] Herein, we wish to report a successful result
involving novel tripodal oxazoline receptors.
Results and Discussion
[a] Prof. Dr. K. H. Ahn, Dr. S.-G. Kim
Department of Chemistry and Division of Molecular & Life Science
Pohang University of Science and Technology
San 31 Hyoja-dong, Pohang 790-784(South Korea)
Fax : (‡82) 54-279-3399
We synthesized various tris(oxazolines) 1 ± 6 with different
substituents and studied the host ± guest interactions (Scheme 1).
Since all receptors are insoluble in water and the ammo-
nium salts are also insoluble in the organic layer in the
absence of receptors, we determined the association constant
E-mail: ahn@postech.ac.kr
Chem. Eur. J. 2000, 6, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0618-3399 $ 17.50+.50/0
3399

S.-G. Kim and K. H. Ahn
FULL PAPER
that has phenyl substituents on oxazoline rings (logKass
ˆ
R2
R1
6.65 Æ 0.02mÀ1). The receptor 6 that has no substituent on
1 R1 = H, R2 = iPr
2 R1 = R2 = Me
3 R1 = Me, R2 = H
4 R1 = H, R2 = Ph
5 R1 = H, R2 = Bn
6 R1 = R2 = H
O
N
oxazoline rings also exhibits a similar level of affinity toward
‡
O
nBuNH3 , whereas those with isopropyl or dimethyl sub-
N
stituents, oxazolines 1 and 2, show much lower affinities. The
oxazoline 3 that has methyl substituent is a borderline case.
These results apparently indicate that the steric interactions
involving the oxazoline substituents significantly affect the
binding affinity. In addition, hydrophobic interactions play a
role in the case of receptor 4, in which phenyl substituents of
oxazoline rings encompass the guest, thereby providing a
ªhydrophobic wallº.[9] Notably, the oxazoline receptor 5
having benzyl substituents exhibits much lower affinities than
the phenyl-substituted oxazoline 4. The benzyl substituents
may not provide the hydrophobic wall but exert steric strain
toward the guests. Likewise, sterically more demanding
alkylammonium ions will disfavor complexation than steri-
1 R2
R1
R
N
R2
O
O
Cl
O
R1
HO2C
R1
R1
O
CO2H
a
Cl
R1
Cl
R1
R1
HO2C
b
R4
R3
R2
HO
‡
‡
NH
O
cally less demanding nBuNH3 . In the case of tBuNH3 , a
much lower affinity is observed with receptor 4. The
R1
H
R2
R3
R4 OH R1
N
c, d
‡
‡
Tris(oxazolines) 1-9
nBuNH3 /tBuNH3 selectivity of as high as ꢀ700 is obtained.
O
R1
O
‡
The binding affinity toward nBuNH3 and the selectivity ratio
OH
obtained with receptor 4 are comparable to those obtained
with the calix[5]arene-based receptor[7] that has been the most
efficient one reported to date. It is noteworthy that oxazoline
6 exhibits a high affinity (Kass ꢀ105 mÀ1) toward sterically
HN
R2
R4
R3
Scheme 1. Synthesis of tris(oxazolines). a) (COCl)2, cat. DMF, CH2Cl2,
258C; b) amino alcohol, Et3N, CH2Cl2, 258C; c) SOCl2, benzene, 258C;
d) NaOH, MeOH, reflux.
‡
demanding tBuNH3 . Oxazoline 4 also shows a significant
affinity toward sBuNH3 (Kass ꢀ105 mÀ1), which is a prereq-
‡
uisite for high chiral recognition.[10] Besides, we studied the
binding affinity toward b-phenethylammonium ion that is
structurally related to biogenic amines such as dopamine and
serotonin. Interestingly, with oxazoline receptors 4 and 6, we
could not correctly determine the association constants due to
large fluctuations in the observed data. For the b-phenethyl-
ammonium ion, even receptor 3 exhibits an association
constant as high as 106 mÀ1. It is likely that two receptors have
very strong affinities toward the ammonium ion (Kass
ꢀ108 mÀ1) that makes the direct measurement unreliable.
Both receptors 4 and 6 extracted b-phenethylammonium ion
almost completely, which is very promising for the possible
application to the detection of dopamine and its analogues.
The benzene-based tripodal oxazolines are in fast equili-
brium between the two conformers, syn,syn- and syn,anti
isomers, in which the relative positioning of three oxazoline
ligands are different.[11] Receptors that are capable of
favorable preorganization are expected to enhance the bind-
ing affinity.[12] Tripodal oxazolines based on 2,4,6-triethylben-
zene frame are synthesized to see this point. Indeed,
tris(oxazoline) 8 that differs from 4 in the frame show a
significant increase in both the binding affinity and selectivity,
as listed in Table 2.
by the picrate extraction method.[8] Table 1 summarizes the
association constants determined with tris(oxazolines) 1 ± 6
having 2,4,6-trimethylbenzene framework.
For a given butylammonium ion, a dramatic change in the
binding affinity was observed. Within the series, the largest
‡
association constant for nBuNH3 is marked by receptor 4
Abstract in Korean:
R3
R2
O
N
R1
7 R1 = Me, R2 = R3 = Ph
8 R1 = Et, R2 = Ph, R3 = H
9 R1 = Et, R2 = R3 = Ph
O
R3
N
R1
R1
R2
N
R3
O
R2
3400
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0618-3400 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 18

Artificial Receptors
3399±3403
Table 1. Association constants (logKass Æ s) of alkylammonium ions (picrate salts) with oxazoline 1 ± 6, obtained
the cation ± p interactions and
an unambiguous evidence for
the hydrophobic wall suggested
previously. Figure 2 shows spec-
tral changes on complexation.
Particularly notable are that
the benzylic protons, more ap-
parently the methyl protons, of
by UV spectroscopy.[a]
‡
‡
‡
‡
‡
‡[b]
Receptor
nBuNH3
sBuNH3
tBuNH3
Ph(CH2)2NH3
nBuNH3 /tBuNH3
1
2
3
4
5
6
4.40 Æ 0.05
4.40 Æ 0.01
5.28 Æ 0.08
6.65 Æ 0.02
4.81 Æ 0.01
6.41 Æ 0.03
3.53 Æ 0.05
3.28 Æ 0.02
4.20 Æ 0.02
6.65 Æ 0.02
3.86 Æ 0.01
5.60 Æ 0.02
3.00 Æ 0.01
3.04 Æ 0.01
3.70 Æ 0.01
3.80 Æ 0.01
3.23 Æ 0.03
5.26 Æ 0.02
4.67 Æ 0.08)
4.38 Æ 0.0423
6.61 Æ 0.02
[c]
25
38
710
38
5.59 Æ 0.09
[c]
14
the
2,4,6-trimethylbenzene
[a] Measured in CHCl3 at 258C (deviation of minimum two independent determinations). [b] Selectivity factor.
[c] Could not be determined due to very large deviations.
frame exhibit upfield shifts
and that their magnitudes de-
pend on the binding affinity.
Table 2. Association constants (logKass Æ s) of alkylammonium ions (pic-
The upfield shift (ÀDd) of the methyl protons on complex-
ation is 0.03 for tris(oxazoline) 2, whereas those for tris(oxa-
zoline) 6 and 4 are 0.15 and 0.14, respectively. Thus, larger
upfield shifts are observed for the complexes having larger
association constants. One likely explanation for the upfield
rate salts) with oxazoline 7 ± 9, obtained by UV spectroscopy.[a]
‡
‡
‡
‡
‡[b]
Receptor nBuNH3
sBuNH3
tBuNH3
nBuNH3 /tBuNH3
7
8
9
6.30 Æ 0.0.9424
Æ 0.05 3.67 Æ 0.01
430
7.96 Æ 0.08 5.94 Æ 0.05 4.53 Æ 0.01 2700
7.77 Æ 0.02 5.92 Æ 0.03 4.17 Æ 0.01 4000
[a] Measured in CHCl3 at 258C (deviation of minimum two independent
determinations). [b] Selectivity factor.
The association constant observed by tris(oxazoline) 8
toward nBuNH3 approaches 108 mÀ1
!
Furthermore, the
‡
[13]
‡
‡
selectivity ratio of nBuNH3 /tBuNH3 is increased to 2700.
This selectivity is even more enhanced to 4000 with hexa-
phenyl-substituted tris(oxazoline) 9. By changing the benzene
‡
frame, the binding affinity toward BuNH3 is raised by up to
20 times. These results demonstrate again the importance of
preorganization in the molecular recognition process.
Regarding to the binding mode, the X-ray crystal structure
‡
À
of 4 ´ PhCH2CH2NH3 PF6 shown in Figure 1 exhibits the
tripodal hydrogen bonding.[14] In addition to the hydrogen
Figure 2. 1H-NMR spectra (300 MHz, CDCl3/CD3OD 9:1, 258C) of
tris(oxazoline) receptors (2: A, 6: C, and 4: E) and their 1:1 complexes
‡
of nBuNH3 picrate (B, D, and F).
‡
À
Figure 1. The X-ray crystal structure of 4 ´ PhCH2CH2NH3 PF6 (except
‡
for the NH3 hydrogen atoms, hydrogen atoms have been omitted for
shift is that the cation ± p interaction perturbs the delocaliza-
tion of the benzene p electrons, thus causing a decrease in
their deshielding effects.[16] Another interesting point is that
the alkyl protons of the guest exhibit dramatic upfield shifts in
the case of receptor 4, compared with the other cases. For
clarity). Selected interatomic distances [Š] and angles [8]: N(1G) ´´´ N(1)
2.972, N(1G) ´´´ N(2) 3.016, N(1G) ´´´ N(3) 2.983; N(1)-N(1G)-N(2) 124.1,
N(1)-N(1G)-N(3) 109.9, N(2)-N(1G)-N(2) 109.6.
‡
bonds, there exists cation ± p interactions between the ammo-
nium ion and the benzene framework that contribute to the
stabilization of the complex.[15]
1H NMR spectroscopic studies were performed on 1:1
complexes of nBuNH3 with the tris(oxazolines) in CD3OD/
CDCl3 1:9, which provided an interesting result with regard to
example, a-methylene protons of nBuNH3 appear at d ˆ 0.87
in the case of 4, whereas those in the case of receptors 2 and 6
appear at d ˆ 2.76 and 2.62, respectively. The dramatic upfield
shifts in the case of receptor 4 may be due to the shielding
effects by the oxazoline phenyl groups that surround the alkyl
chain to form a hydrophobic wall.
‡
Chem. Eur. J. 2000, 6, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0618-3401 $ 17.50+.50/0
3401

S.-G. Kim and K. H. Ahn
FULL PAPER
zolines 1 and 3 can be synthesized by a different route reported in the
reference [1].
Conclusion
1,3,5-Tris[(4,4-dimethyl-2-oxazolinyl)methyl]-2,4,6-trimethylbenzene (2):
Overall yield 45%; Rf ˆ 0.21 (ethyl acetate/MeOH 95:5 v/v); m.p. 184±
1858C; 1H NMR (300 MHz, CDCl3): d ˆ 3.87 (s, 6H), 3.69 (s, 6H), 2.37 (s,
9H), 1.24(s, 18H); 13C NMR (75 MHz, CDCl3): d ˆ 164.4, 136.2, 131.0,
We have developed the benzene-based tripodal tris(oxazo-
lines) as the most selective and strong receptors toward linear
alkylammonium ions reported to date. The X-ray crystallo-
graphic and 1H NMR spectroscopic studies provide unambig-
uous binding modes both in the solid and in solution states.
The new receptor system has several unique features such as
ready availability, structural simplicity, and, in particular,
versatility in derivatization. By virtue of these advantages, it
can be readily tailored as selective receptors toward bio-
logically important amines.
‡
79.5, 67.2, 30.7, 28.7, 17.4; MS (EI): m/z (%): 453 (100) [M] , 382 (51), 310
(37); elemental analysis calcd (%) for C27H39N3O3 ´ 1ꢁ2H2O: C 70.10 H 8.72,
N 9.08; found: C 70.35, H 8.33, N 9.09.
1,3,5-Tris{[4(S)-phenyl-2-oxazolinyl]methyl}-2,4,6-trimethylbenzene (4):
Overall yield 28%; Rf ˆ 0.34(ethyl acetate/hexanes 3:2 v/v); m.p. 125 ±
1268C; [a]2D7 ˆ ‡77.8 (c ˆ 1.00, CHCl3); 1H NMR (300 MHz, CDCl3): d ˆ
7.34± 7.21 (m, 15H), 5.16 (dd, J ˆ 10.1, 8.3 Hz, 3H), 4.59 (dd, J ˆ 10.1,
8.5 Hz, 3H), 4.05 (dd, J ˆ 8.5, 8.3 Hz, 3H) 3.87 (s, 6H), 2.51 (s, 9H);
13C NMR (75 MHz, CDCl3): d ˆ 167.0, 142.5, 135.9, 130.7, 128.6, 127.4,
‡
126.5, 74.8, 69.5, 30.1, 17.3; MS (EI): m/z (%): 597 (100) [M] , 576 (28);
elemental analysis calcd (%) for C39H39N3O3: C 78.36, H 6.58, N 7.03;
found: C 78.13, H 6.70, N 7.41.
Experimental Section
1,3,5-Tris{[4(S)-phenylmethyl-2-oxazolinyl]methyl}-2,4,6-trimethylben-
zene (5): Overall yield 30%; Rf ˆ 0.24(ethyl acetate/hexanes 3:2 v/v); m.p.
139 ± 1408C; [a]2D7 ˆ À16.7 (c ˆ 1.00, CHCl3); 1H NMR (300 MHz, CDCl3):
d ˆ 7.30 ± 7.16 (m, 15H), 4.38 ± 4.30 (m, 3H), 4.12 (t, J ˆ 8.3 Hz, 3H) 3.94(t,
J ˆ 7.7 Hz, 3H), 3.71 (s, 6H), 3.09 (dd, J ˆ13.7, 4.7 Hz, 3H), 2.6 (dd, J ˆ 13.7,
8.9 Hz, 3H), 2.38 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 166.6, 138.3,
136.2, 131.1, 129.7, 128.9, 126.8, 72.0, 67.6, 42.0, 30.5, 17.6; MS (EI): m/z (%):
General: Benzene and dichloromethane were purified by distillation from
calcium hydride under nitrogen. Thionyl chloride was distilled under argon
before use. Alkylammonium picrates were prepared by neutralization of
the appropriate amine with picric acid in methanol and purified by
recrystallization from diethyl ether/methanol. Analytical thin layer chro-
matography was performed using Merck 60 F254 precoated silica gel glass
plates. Column chromatography was carried out on Merck silica gel 60
(230 ± 400 mesh). Melting points were measured on a Thomas Hoover
capillary melting point apparatus and were uncorrected. Optical rotations
were measured using Rudolph Research Autopol III digital polarimeter
using a sodium lamp (D line, 589 nm) and are reported in degrees with
concentration in unit of 10 mgmLÀ1. Proton and carbon NMR spectra were
recorded on an AM-300 Bruker spectrometer. Chemical shifts are reported
as d in ppm downfield from tetramethylsilane (d ˆ 0.0) using the residual
solvent signal as an internal standard: [D1]chloroform (1H NMR: d ˆ 7.26;
13C NMR: d ˆ 77.0). Multiplicity is given as usual. Mass spectral analysis
was recorded on Jeol JMS-AX505WA and is reported in units of mass to
charge (m/z). Elemental analyses and HR-MS were performed by Seoul
Branch Analytical Laboratory of Korea Basic Science Institute.
‡
639 (82) [M] , 548 (100); elemental analysis calcd (%) for C42H45N3O3: C
78.84, H 7.09, N 6.57; found: C 79.02, H 7.46, N 6.55.
1,3,5-Tris[(2-oxazolinyl)methyl]-2,4,6-trimethylbenzene (6): overall yield
1
44%; Rf ˆ 0.43 (ethyl acetate/MeOH 4.1 v/v); m.p. 285 ± 2908C; H NMR
(300 MHz, CDCl3): d ˆ 4.17 (t, J ˆ 9.4Hz, 3H), 3.75 (t, J ˆ 9.4Hz, 3H) 3.66
(s, 6H), 2.34(s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 167.0, 136.0, 131.1,
‡
67.8, 54.8, 30.3, 17.4; MS (EI): m/z (%): 369 (100) [M] , 326 (34), 283 (32);
HR-MS (EI): calcd for C21H27N3O3: 369.2052; found: 369.2052.
1,3,5-Tris{[(S,S)-4,5-diphenyl-2-oxazolinyl]methyl}-2,4,6-trimethylbenzene
(7): Prepared similarly with (1R,2S)-2-amino-1,2-diphenylethanol in over-
all yield 48%; Rf ˆ 0.58 (ethyl acetate/hexanes 1:1 v/v); m.p. 148 ± 1498C;
[a]2D5 ˆ À107.2 (c ˆ 0.50, CHCl3); 1H NMR (300 MHz, CDCl3): d ˆ 7.37 ±
7.20 (m, 30H), 5.21 (d, J ˆ 7.3 Hz, 3H), 5.04(d, J ˆ 7.3 Hz, 3H), 4.06 (s, 6H)
2.69 (s, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 167.1, 142.7, 141.3, 136.4,
131.4, 129.4, 129.3, 128.8, 128.1, 127.1, 126.0, 89.4, 79.1, 30.9, 18.1; MS
Synthesis of the tris(oxazolines) 1 ± 7ÐA representative procedure: Oxalyl
chloride (4.3 mL, 50 mmol) and dimethylformamide (0.2 mL, 2.5 mmol)
were added at 258C to a mixture of (2,4,6-trimethylbenzene)-1,3,5-triacetic
acid (10 mmol) in dichloromethane (140 mL). After the reaction mixture
was stirred for 18 h, the solvent and excess oxalyl chloride were evaporated
under reduced pressure. The crude acid chloride was immediately used for
the next step without further purification. A solution of the acid chloride in
dichloromethane (40 mL) was added dropwise through a cannula to a
stirred solution of an amino alcohol (30 mmol) and triethylamine
(4.7 mmol, 34 mmol) in dichloromethane (50 mL) at 08C. After the
solution was stirred for 12 h at 258C, the solvent was evaporated under
reduced pressure to give the corresponding crude triamide. To the triamide
were added benzene (140 mL) and thionyl chloride (7.2 mL, 100 mmol),
and the resulting mixture was vigorously stirred at 258C for 18 h. After
evaporating excess thionyl chloride and solvent under reduced pressure,
the residue was dissolved in dichloromethane/water (200 mL, 1:1 v/v) and
then stirred vigorously for 30 min at 258C. The organic layer was separated,
and the aqueous layer was extracted with dichloromethane. The combined
organic layers were washed with brine, dried over MgSO4, and concen-
trated in vacuo to afford the crude chloromethylamide as a light yellow
solid, which was subjected to the next step without further purification. A
solution of sodium hydroxide (4.0 g, 100 mmol) in methanol (40 mL)
through a cannula was added to a mixture of chloromethylamide dissolved
in methanol (100 mL). The resulting mixture was stirred for 30 min at 258C
and then heated under reflux for 12 h. After being cooled to room
temperature, the mixture was concentrated under reduced pressure. The
residue was re-dissolved in dichloromethane (20 mL) and neutralized with
1n HCl solution, washed with brine, and dried over MgSO4. After
concentration in vacuo, the crude product was purified by column
chromatography on SiO2 and then by recrystallization to afford an
analytically pure compound. Similarly, tris(oxazolines) 8 and 9 can be
synthesized starting from (2,4,6-triethylbenzene)-1,3,5-triacetic acid. Oxa-
‡
(FAB): m/z (%): 826 (58) [M] , 307 (100); elemental analysis calcd (%)
C57H51N3O3: C 82.88, H 6.22, N 5.09; found: C 82.66, H 6.48, N 5.06.
1,3,5-Tris{[4(S)-phenyl-2-oxazolinyl]methyl}-2,4,6-triethylbenzene
(8):
overall yield 51%; Rf ˆ 0.25 (ethyl acetate/hexanes 1:1 v/v); m.p. 63 ±
648C; [a]2D5 ˆ À41.4 (c ˆ 0.50, CHCl3); 1H NMR (300 MHz, CDCl3): d ˆ
7.32 ± 7.20 (m, 15H), 5.15 (t, J ˆ 9.3 Hz, 3H), 4.58 (dd, J ˆ 10.2, 8.4Hz, 3H),
4.04 (t, J ˆ 8.4Hz, 3H) 3.85 (s, 6H), 2.96 (dd, J ˆ 7.5 Hz, 6H), 1.21 (t, J ˆ
7.5 Hz, 9H); 13C NMR (75 MHz, CDCl3): d ˆ 168.2, 142.9, 142.7, 130.2,
129.0, 127.8, 127.0, 75.3, 69.9, 29.2, 23.8, 15.1; MS (FAB): m/z (%): 640 (100)
‡
‡
[M] , 307 (61); HR-MS (FAB): calcd for C42H45N3O3 [M‡H] : 640.3539;
found: 640.3544.
1,3,5-Tris{[(S,S)-4,5-diphenyl-2-oxazolinyl]methyl}-2,4,6-triethylbenzene
(9): overall yield 59%; Rf ˆ 0.65 (ethyl acetate/hexanes 1:1 v/v); m.p. 81 ±
828C; [a]2D5 ˆ À107.4( c ˆ 0.50, CHCl3); 1H NMR (300 MHz, CDCl3): d ˆ
7.24± 7.05 (m, 30H), 5.04(d, J ˆ 7.6 Hz, 3H), 4.93 (d, J ˆ 7.6 Hz, 3H), 3.99
(s, 6H), 3.14(q, J ˆ 7.4Hz, 6H), 1.28 (t, J ˆ 7.4Hz, 9H); 13C NMR
(75 MHz, CDCl3): d ˆ 167.8, 143.2, 142.7, 130.6, 129.4, 128.7, 128.2, 127.3,
126.6, 126.1, 125.5, 89.4, 79.2, 29.7, 24.0, 15.6; MS (FAB): m/z (%): 868 (100)
‡
[M] , 761 (9), 307 (52); elemental analysis calcd (%) for C60H57N3O3: C
83.01, H 6.62, N 4.84; found: C 82.69, H 6.87, N 4.80.
Determination of Kass: A mixture of the picrate salt (0.5 mL, 0.015m in
water) and the host (0.1 mL, 0.075m in CHCl3) in a centrifuge tube was
equilibrated for 1 h in a thermostat at 258C. After being kept for 1 h, the
whole mixture was extracted by Vortex ± Genie for 1 min and then
centrifuged at 1500 rpm for 1 min. An aliquot of the CHCl3 layer was
measured and transferred by micro-syringe into a 5 mL volumetric flask
and diluted to the mark with CH3CN. For a more intensely colored layer
0.01 mL aliquot and for a less intensely colored layer 0.05 mL aliquot were
used. The UV absorption of each 5 mL solution was measured at 380 nm
3402
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0618-3402 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 18

Artificial Receptors
3399±3403
using Perkin ± Elmer HP 8452 UV/Vis spectrometer. Calculation of the
binding constant was followed by the literature procedure.[8]
b) C. M. Hartshorn, P. J. Steel, J. Chem. Soc. Chem. Commun. 1997,
541 ± 542; c) B. F. Hoskins, R. Robson, D. A. Slizys, Angew. Chem.
1997, 109, 2861 ± 2863; Angew. Chem. Int. Ed. Engl. 1997, 36, 2752 ±
2754; d) A. Metzger, V. M. Lynch, E. V. Anslyn, Angew. Chem. 1997,
109, 911 ± 914 Angew. Chem. Int. Ed. Engl. 1997, 36, 862 ± 865; e) C.
Walsdorff, W. Saak, S. Pohl, J. Chem. Soc. Dalton Trans. 1997, 1857 ±
1861; f) Z. Hou, T. D. P. Stack, C. J. Sunderland, K. N. Raymond,
Inorg. Chim. Acta 1997, 263, 341 ± 355; g) K. Niikura, A. Metzger,
E. V. Anslyn, J. Am. Chem. Soc. 1998, 120, 8533 ± 8534; g) T. Szabo,
B. M. OꢂLeary, J. Rebek, Jr. Angew. Chem. 1998, 110, 3606 ± 3609;
Angew. Chem. Int. Ed. 1998, 37, 3410 ± 3413; h) K. Sato, S. Arai, T.
Yamagishi, Tetrahedron Lett. 1999, 40, 5219 ± 5222; i) L. A. Cabell, M.-
K. Monahan, E. V. Anslyn, Tetrahedron Lett. 1999, 40, 7753 ± 7756.
[12] For preorganization effects, see: a) J. M. Lehn, Supramolecular
Chemistry, Concepts and Perspective, VCH, Weinheim, 1995;
b) W. C. Still, Acc. Chem. Res. 1996, 29, 155 and references therein.
[13] Because of lack of data, the affinity can not be compared; however,
tris(oxazoline) 8 is believed to be the most strong receptor toward
alkylammonium ions in which only weak binding forces interplay. This
unusual affinity is attributed to the ideal tripodal hydrogen bonding,
cation ± p interactions, and hydrophobic effects.
Acknowledgement
This work was supported by the Center for Biofunctional Molecules,
sponsored by Korea Science and Engineering Foundation and POSTECH
research fund. We thank Prof. S. K. Shin for helpful discussion. We also
thank K.-H. Kim and J. Jung for their assistance in the synthesis of the
oxazoline receptors.
[1] K. H. Ahn, S.-G. Kim, J. Jung, K.-H. Kim, J. Kim, J. Chin, K. Kim,
Chem. Lett. 2000, 170 ± 171.
‡
‡
[2] The selective recognition of NH4 over K by benzene-based tripodal
pyrazole receptors has been recently reported: J. Chin, C. Walsdorff,
B. Stranix, J. Oh, H. J. Chung, S.-M. Park, K. Kim, Angew. Chem. 1999,
111, 2923 ± 2926; Angew. Chem. Int Ed. Engl. 1999, 38, 2756 ± 2759.
[3] D. J. Cram, J. M. Cram, Acc. Chem. Res. 1978, 11, 8 ± 14 .
[4] P. Bühlmann, E. Pretsch, E. Bakker, Chem. Rev. 1998, 98, 1593 ± 1687.
[5] a) P. Krogsgaard-Larsen in Comprehensive MedicinalChemistry , Vol.
3 (Eds.: P. G. Sammes, J. B. Taylor, J. C. Emmett), Pergamon, Oxford,
1990, pp. 493 ± 537; b) A. S. Horn in Comprehensive MedicinalChem-
[14] Crystal data for 4 ´ C8H9NH3PF6 ´ CH2Cl2: 0.25 Â 0.25 Â 0.25 mm
C48H53Cl2F6N4O3P, Mr ˆ 949.81, monoclinic, space group P21, a ˆ
11.8497(5), b ˆ 17.5041(7), c ˆ 12.6934(6) Š, a ˆ 908, b ˆ 116.465(1)8,
g ˆ 908, Vˆ 2356.94(18) Š3, Z ˆ 2, Tˆ 223(2) K, 1calcd ˆ 1.337 gcmÀ3
,
istry, Vol.
3 (Eds.: P. G. Sammes, J. B. Taylor, J. C. Emmett),
Siemens SMART CCCD diffractometer, MoKa radiation, 9582
reflections collected, 5795 independent reflections, R1 ˆ 0.0461,
wR2 ˆ 0.0921 [I > 2s(I)], R1 ˆ 0.0594, wR2 ˆ 0.1013 (all data),
GOF ˆ 1.136. Crystallographic data (excluding structure factors) for
the structure reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publica-
tion no. CCDC-147263. Copies of the data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge
CB21EZ, UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.
ac.uk).
Pergamon, Oxford, 1990, pp. 229 ± 290; for biomimetic recognition
and related ammonium ions, see: c) T. Schrader, Angew. Chem. 1996,
108, 2816 ± 2818; Angew. Chem. Int. Ed. Engl. 1996, 35, 2649 ± 2651;
d) T. Schrader, J. Org. Chem. 1998, 63, 264± 272; e) M. Herm, T.
Schrader, Chem. Eur. J. 2000, 6, 47 ± 53.
[6] a) S.-K. Chang, M. J. Jang, S.-Y. Han, Chem. Lett. 1992, 1937 ± 1940;
b) S.-Y. Han, M.-H. Kang, Y.-E. Jung, S.-K. Chang, J. Chem. Soc.
Perkin Trans. 2 1994, 835 ± 839.
[7] F. Arnaud-Neu, S. Fuangswasdi, A. Notti, S. Pappalardo, M. F. Parisi,
Angew. Chem. 1998, 110, 120 ± 122; Angew. Chem. Int. Ed. 1998, 37,
112 ± 114.
[15] An ab initio calculation is carried out by colleagues and will be
reported independently. For
a recent review on the cations-p
[8] G. M. Lein, D. J. Cram, J. Am. Chem. Soc. 1985, 107, 448 ± 455.
[9] For related secondary hydrophobic interactions by the side chain
aromatic rings, see: J.-P. Behr, J.-M. Lehn, P. Vierling, J. Chem. Soc.
Chem. Commun. 1976, 621 ± 623.
[10] Enantiomeric recognition by the C3 symmetric receptors is ongoing
and will be reported in due course.
[11] For recent selected examples on the use of other benzene-based
tripodal ligands in molecular recognition, metal complexation, and
self-assembly, see: a) C. M. Hartshorn, P. J. Steel, Angew. Chem. 1996,
108, 2818 ± 2820; Angew. Chem. Int. Ed. Engl. 1996, 35, 2655 ± 2657;
interactions, see: J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97,
1303 ± 1324.
[16] The effect of the cation ± p interactions on 1H-NMR shifts of the
benzene or benzylic protons has not reported yet. For an interesting
case where a down-field shift is observed in a presumed cation ± p
interaction system, see: S. L. De Wall, E. S. Meadows, L. J. Barbour,
G. W. Gokel, J. Am. Chem. Soc. 1999, 121, 5613 ± 5614.
Received: February 25, 2000 [F2320]
Chem. Eur. J. 2000, 6, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0618-3403 $ 17.50+.50/0
3403