Tuning and dissecting electronic and steric effects in ammonium receptors: Nonactin vs artificial receptors

Article abstract of DOI:10.1021/ja0174175

Ion selective electrodes (ISE) based on three different tripodal receptors (5, 6, and 7) have been investigated for sensing ammonium ion. Each receptor is based on three pyrazole groups that can accept three H-bonds from the bound ammonium ion. The receptor based on 4-bromo-3,5-dimethylpyrazole (6) is the most sensitive with a detection limit for ammonium ion of 2.5 × 10^-5 M at pH 8. The detection limits for the receptors based on 2,3-dimethylpyrazole (5) and unsubstitued pyrazole (7) are 1.0 × 10^-4 and 2.0 × 10^-4 M, respectively. The selectivities of the receptors 5, 6, and 7 for sensing ammonium ion over potassium ion (logK_NH4+/K+) are -2.8, -2.3, and -1.7, respectively. In contrast, the detection limit and the selectivity of a nonactin-based ISE are 2.2 × 10^-5 M and -1.3, respectively. Crystallographic studies reveal that 6 accepts three H-bonds from the bound ammonium and singly protonated receptor 5 forms three H-bonds with the bound water molecule.

Full text of DOI:10.1021/ja0174175

Published on Web 04/19/2002
Tuning and Dissecting Electronic and Steric Effects in
Ammonium Receptors: Nonactin vs Artificial Receptors
Jik Chin,*,† Jinho Oh,‡ Sang Yong Jon,‡ Sang Hyun Park,‡ Christian Walsdorff,‡
Brent Stranix,† Assaad Ghoussoub,† Seok Jong Lee,† Hyun Jei Chung,‡
Su-Moon Park,*,‡ and Kimoon Kim*,‡
Contribution from the Department of Chemistry, UniVersity of Toronto, 80 St. George Street,
Toronto, Ontario, Canada M5S 1A1, National CreatiVe Research InitiatiVe Center for Smart
Supramolecules, Center for Integrated Molecular Systems, and Department of Chemistry,
DiVision of Molecular and Life Sciences, Pohang UniVersity of Science and Technology,
San 31 Hyojadong, Pohang 790-784, Republic of Korea
Received October 29, 2001
Abstract: Ion selective electrodes (ISE) based on three different tripodal receptors (5, 6, and 7) have been
investigated for sensing ammonium ion. Each receptor is based on three pyrazole groups that can accept
three H-bonds from the bound ammonium ion. The receptor based on 4-bromo-3,5-dimethylpyrazole (6) is
the most sensitive with a detection limit for ammonium ion of 2.5 × 10-5 M at pH 8. The detection limits for
the receptors based on 2,3-dimethylpyrazole (5) and unsubstitued pyrazole (7) are 1.0 × 10-4 and 2.0 ×
10-4 M, respectively. The selectivities of the receptors 5, 6, and 7 for sensing ammonium ion over potassium
+
+
ion (logKNH
) are -2.8, -2.3, and -1.7, respectively. In contrast, the detection limit and the selectivity
/K
4
of a nonactin-based ISE are 2.2 × 10-5 M and -1.3, respectively. Crystallographic studies reveal that 6
accepts three H-bonds from the bound ammonium and singly protonated receptor 5 forms three H-bonds
with the bound water molecule.
Chart 1
Introduction
The study of ammonium ion recognition is of considerable
fundamental and practical interest. It is also central to the
understanding of chiral and achiral alkylammonium recognition
(e.g., amino acids and the dopamine class of neurotransmitters).1
The early finding that 18-crown-6 (1)2 is an excellent receptor
for NH4 led to the development of chiral analogues (2)3 for
+
stereospecific recognition of alkylammonium ions (Chart 1). A
cryptand (3)4 has been shown to be even better than crown ethers
for binding NH4+. Nonactin (4), a naturally occurring antibiotic
that binds tightly to NH4+, is currently used in commercial ion
selective electrodes (ISEs) for detecting the cation down to
micromolar levels.5 Such sensors are useful for indirectly
measuring the concentrations of urea or creatine in biological
samples,6 as well as for directly measuring the concentrations
of NH4+ in the environment.7 A major problem with nonactin-
* Corresponding author. E-mail: jchin@chem.utoronto.ca.
† University of Toronto.
based ISEs is that although they are highly sensitive, their
+
selectivity for binding NH4 over K+ is low.
‡ Pohang University of Science and Technology.
In a recent communication, we showed that 5 is over 10 times
more selective than nonactin for binding NH4+ over K+.8 Since
K+ prefers to be coordinated by six or more atoms, we have
been able to increase the selectivity by lowering the number of
coordinating atoms in the receptor from eight in nonactin (4)
to three in 5. However, this approach led to a drastic reduction
(1) Zhang, X. X.; Bradshaw, J. S.; Izatt, R. M. Chem. ReV. 1997, 97, 3313-
3361.
(2) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017-7036.
(3) Cram, D. J. Science 1988, 240, 760-767.
(4) Lehn, J. M. Acc. Chem. Res. 1978, 11, 49-57.
(5) Young, C. C. J. Chem. Educ. 1997, 74, 177-182. Ghauri, M. S.; Thomas,
J. S. R. Analyst 1994, 119, 2323-2326. Beer, D. De.; van den Heuvel, J.
C. Talanta 1988, 35, 728-730. Davies, O. G.; Moody, G. J.; Thomas, J.
D. R. Analyst 1988, 113, 497-500. Ma, S. C.; Chaniotakis, N. A.
Meyerhoff, M. E. Anal. Chem. 1988, 60, 2293-2299.
(6) Wolfbeis, O. S.; Li, H. Biosens. Biolectron. 1993, 8, 161-166.
(7) Buhlman, P.; Pretsch, E.; Bakker, E. Chem. ReV. 1998, 98, 1593-1689.
(8) Chin, J.; Walsdorff, C.; Stranix, B.; Oh, J.; Chung, H. J.; Park, S.-M.; Kim,
K. Angew. Chem., Int. Ed. Engl. 1999, 38, 2756-2759.
9
5374
J. AM. CHEM. SOC. 2002, 124, 5374-5379
10.1021/ja0174175 CCC: $22.00 © 2002 American Chemical Society

Electronic and Steric Effects in Ammonium Receptors
A R T I C L E S
Chart 2
Table 1. Crystal and Structure Refinement Data for
- a
- a
[5H+-H2O] ClO4
and [6‚NH4] PF6
[5H ‚H O] ClO -
[6‚NH ] PF -
+
2
4
4
6
empirical formula
formula wt
crystal size, mm3
crystal system
space group
temp, K
a, Å
C30H43Br3F6 N7P
605.17
0.6 × 0.5 × 0.4
monoclinic
P21/c
C30H43Br3F6 N7P
886.41
0.40 × 0.40 × 0.25
triclinic
P1h
188(2)
243(2)
8.9653(1)
19.4149(3)
18.4104(2)
90
90.086(1)
90
3204.52(7)
4
1.254
0.166
12615
4967
403
10.8201(2)
11.8525(3)
16.0993(1)
93.654(1)
98.994(1)
113.193(1)
1856.53
2
1.586
3.364
7581
5567
b, Å
c, Å
R, deg
â, deg
γ, deg
vol, Å3
Z
d
calc, g/cm3
µ, mm-1
no. of reflcns collected
no. of unique reflcns
no. of variables
R1, wR2, %
428
5.92, 15.49
1.055
+
4.45, 12.23
0.764
in the sensitivity of the receptor for detecting NH4 at neutral
pH. Here we investigate how steric and electronic effects
influence the selectivity and sensitivity of the receptors by
comparing the performances of ISEs based on 5, 6, and 7 (Chart
goodness of fit
a All the data were collected on a SMART CCD diffractometer with
Mo KR radiation (λ ) 0.71071 Å).
2). We also report the effect of solution pH on the sensitivity
(6H, q, CH2), 0.90 (9H, t); 13C NMR (75 MHz, CDCl3) δ 146.44,
139.67, 130.81, 128.31, 105.99, 49.96, 23.70, 15.58.
+
of the receptors for detecting NH4
.
Complex [6‚NH4] PF6-: The complex was generated by mixing
equimolar quantities of 6 and NH4PF6 in CH3CN/CHCl3 (1:1 v/v). 1H
NMR (CDCl3, 300 MHz) δ 6.60(4H, br) 5.17(6H, s, CH2), 2.50(6H,
br, CH2), 2.30(9H, br, CH3), 2.11(9H, br, CH3), 1.23(9H, br, CH3);
FAB-MS m/z 739 (M+). Anal. Calcd for C30H45Br3F6N7OP: C, 39.84;
H, 5.02, N, 10.84. Found: C, 39.79; H, 4.82; N, 10.75.
Experimental Section
Instruments. 1H NMR and 13C NMR spectra were recorded on
Bruker DPX 300 and DRX 500 NMR spectrometers at room temper-
ature. Chemical shifts are reported in ppm with tetramethylsilane used
as reference.
Materials. All commercially available reagents were purchased from
Aldrich or Fluka and used without further purification. Tetrahydrofuran
was dried by distillation from sodium metal.
Compound 5: To a suspension of NaH (0.62 g, 16 mmol) in 40
mL of anhydrous THF was added 3,5-dimethylpyrazole (1.48 g, 16
mmol) in small portions at room temperature. The mixture was stirred
for 20 min until the evolution of hydrogen gas ceased. The resulting
solution was slowly added to a slurry of 1,3,5-tris(bromomethyl)-2,4,6-
triethylbenzene (2.0 g, 4.6 mmol) in 20 mL of anhydrous THF. After
being stirred for 3 h, the mixture was poured into water (50 mL) and
extracted with CH3Cl (50 mL × 2). The combined organic extract was
dried over MgSO4 and evaporated under reduced pressure. The
remaining solid was washed with ether to remove excess 3,5-
dimethylpyrazole and dried in vacuo to give 1.62 g (73%) of the product
as colorless crystals. 1H NMR (CDCl3, 300 MHz) δ 0.86 (9 H, t, Me),
2.14 (9 H, s, Me), 2.15(9 H, s, Me), 2.77 (6 H, q, CH2), 5.18 (6 H, s,
CH2), 5.76 (3 H, s, pyrazol-4-yl-H); 13C NMR (CDCl3, 75 MHz) δ
11.85, 13.75, 15.09, 24.12, 47.59, 106.03, 130.83, 139.85, 145.62,
147.38. Anal. Calcd for C30H42N6: C, 74.03; H, 8.70; N, 17.27.
Found: C, 73.79; H, 8.71; N, 17.04.
Complex [5H+‚H2O] ClO4-: The complex was obtained as a white
powder by adding perchloric acid to a solution of 5 dissolved in dilute
hydrochoric acid. Recrystallization of the white powder from ethanol
1
gave colorless rectangular plates of the product. H NMR (CD3OD,
500 MHz) δ 6.20(3H, s, pyrazol-4-yl-H), 5.39 (6H, s, CH2), 2.67 (6H,
q, CH2), 2.51 (9H, s, CH3), 2.20 (9H, s, CH3), 0.95 (9H, t, CH3); 13C
NMR (125 MHz, CD3OD) δ 147.86, 147.31, 143.66, 129.32, 107.06,
46.45, 23.59, 14.06, 11.13, 10.42. Anal. Calcd for C30H45ClN6O: C,
59.54; H, 7.49, N, 13.89. Found: C, 59.13; H, 7.46; N, 13.75.
Crystallography. The intensity data for [5H+‚H2O ] ClO4 and
-
[6‚NH4] PF6- were collected using a Siemens diffractometer equipped
with a graphite-monochromated Mo ΚR (λ ) 0.71073 Å) radiation
source and a CCD detector. The structures were solved by pattern
synthesis (SHELXLS86) and subsequent difference Fourier methods
(SHELXL93). Calculations were performed with the SHLXTL v5.1
program.9 Semiempirical absorption corrections were not applied for
the structures. All the nonhydrogen atoms were refined anisotropically.
While most hydrogen atoms of the complex [5H+‚H2O] ClO4 were
-
generated with ideal geometry, the hydrogen atoms around water (H1,
H2, and H3) were located from difference Fourier maps and refined
without any constraints. Crystal and structure refinement data for
Compound 6: To a solution of 5 (1.0 g, 2.1 mmol) in CHCl3 was
slowly added Br2 (0.42 mL, 8.2 mmol) at 0 °C. The reaction mixture
was stirred for 1 h and poured into a saturated solution of K2CO3 (50
mL) and extracted with CHCl3 (30 mL × 2). The extracts were
-
-
[5H+‚H2O ] ClO4 and [6‚NH4] PF6 are summarized in Table 1.
H-bonding interactions are listed in Tables 2 and 3. The illustrations
of the two crystal structures prepared by using the ORTEP10 program
are shown in Figures 1 and 2.
1
evaporated to give 1.43 g (97%) of the product as a yellow solid. H
NMR (CDCl3, 300 MHz) δ 5.16 (6H, s, CH2), 2.68 (6H, q, CH2), 2.10
(18H, d, CH3), 0.86 (9H, t, CH3); 13C NMR (CDCl3, 75 MHz) δ 146.38,
145.82, 137.47, 130.37, 94.91, 48.26, 24.12, 15.08, 12.70, 10.94. FAB-
MS m/z 721 (M + 1)+. Anal. Calcd for C30H39Br3N6: C, 49.81; H,
5.43; N, 11.62. Found: C, 49.65; H, 5.51; N, 11.62.
General Procedure for Preparation of Ion Selective Electrodes
and Measurement of Their Performances. The ionophore (1 mg),
PVC (33 mg), and bis(2-ethylhexyl)adipate (DOA as a plasticizer, 66
mg) were dissolved in anhydrous THF. The mixture was vigorously
Compound 7: The product was obtained as colorless crystals in
65% yield following the same general procedure for making 5. 1H
NMR(CDCl3, 300 MHz,) δ 7.48 (3H, d, pyrazol-5-yl-H), 6.95 (3H, d,
pyrazol-5-yl-H), 6.15 (3H, t, pyrazol-4-yl-H), 5.39 (6H, s, CH2), 2.66
(9) SHELXTL v5.1 from Bruker-AXS, Inc.: Madison, WI, 1998.
(10) ORTEP: Johnson, C. K. 1976. ORTEP-A Fortran Thermal Ellipsoid Plot
Program; Technical Report ORNL-5138; Oak Ridge National Laboratroy:
Oak Ridge, TN, 1976.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5375

A R T I C L E S
Chin et al.
Table 2. Selected Distances(Å) and Angles (deg) for
[5H+‚H2O] ClO4
-
bond distances
O(1)-H(1)
O(1)-H(3)
H(3)-N(2)
O(1)-O(5)
N(6)-O(1)
N(6)-O(5)
1.78 (3)
0.88 (3)
2.01 (3)
5.47 (1)
2.67 (2)
3.85 (3)
O(1)-H(2)
H(2)-N(4)
H(1)-N(6)
O(5)-N(6)
N(4)-O(1)
0.84 (3)
2.07 (3)
0.92 (3)
3.85 (2)
2.89 (3)
bond angles
167 (3) O(1)-H(3)-N(2)
163 (2)
O(1)-H(2)-N(4)
O(1)-H(1)-N(6)
174 (3)
-
Table 3. Selected Distances(Å) and Angles (deg) for [6‚NH4] PF6
bond distances
N(7)-H(1)
N(7)-H(3)
H(1)-F(6)
H(3)-N(4)
N(7)-F(6)
N(7)-N(4)
0.85 (7)
0.92 (8)
2.37 (7)
2.02 (8)
3.18 (1)
2.92 (1)
N(7)-H(2)
N(7)-H(4)
H(2)-N(6)
H(4)-N(2)
N(7)-N(2)
N(7)-N(6)
0.83 (7)
0.90 (8)
2.09 (8)
2.10 (8)
2.99 (1)
2.91 (1)
bond angles
N(7)-H(1)-F(6)
N(7)-H(3)-N(4)
159 (5)
165 (6)
N(7)-H(4)-N(2)
N(7)-H(2)-N(6)
171 (6)
167 (6)
-
Figure 2. Structure of [6‚NH4+] PF6 ((a) top and (b) side views) with
the thermal ellipsoids set at the 50% probability level.
electrode bodies (IS-561). Each electrode was filled with an ammonium
chloride solution (0.01 M) as an internal reference soaked in distilled
water for 12 h before use. Potential differences between the ISEs and
the Orion sleeve-type double juntion Ag/AgCl reference electrode
(Model 90-02) were recorded using an IBM AT-type computer equipped
with a 16 channel analog-to-digital converter. The dynamic response
curves were obtained at room temperature by adding standard solutions
of ammonium chloride or potassium chloride to 200 mL of water (0.05
M Tris-HCl buffer, pH 7.2, 8.0, and 9.0, respectively) every 100 s to
vary the concentrations of each ionic species stepwise (Figure 3). The
potentials were measured every second at room temperature. Selectivity
coefficients were estimated according to the fixed interference method
(FIM, 0.1 M interference ion).11
Results and Discussion
In designing novel receptors, much can be learned from the
+
studies of nonactin. Crystallographic studies showed that NH4
or K+ is positioned at the center of a cube formed from the
four ether oxygens and the four carbonyl oxygens of nonactin
(4).12,13 The average distance between the eight oxygen atoms
Figure 1. Structure of [5H+‚H2O] ClO4- ((a) top and (b) side views) with
the thermal ellipsoids set at the 50% probability level.
(11) Umezawa, Y.; Umezawa, K.; Sato, H. Pure Appl. Chem. 1995, 67, 507-
518.
stirred for 6 h. The resulting solution was poured into a glass ring (i.d.
22 mm) on a glass slide and dried over a day at room temperature.
Small disks were punched from the cast films and mounted on Philips
(12) Neuport-Laves, K.; Dobler, M. HelV. Chim. Acta 1976, 59, 614-623.
(13) Dobler, M.; Dunitz, J. D.; Kilburn, B. T. HelV. Chim. Acta 1969, 52, 2573-
2583.
9
5376 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

Electronic and Steric Effects in Ammonium Receptors
A R T I C L E S
Scheme 1
H-bonds with the bound H2O molecule while the three ethyl
groups point to the opposite side as expected from molecular
mechanics computations.15,16
+
Figure 2 shows the crystal structure of NH4 bound to 6. It
+
is interesting to compare the NH4 and H2O bound structures
(Figures 1 and 2). In binding NH4+, the neutral receptor (6)
accepts three H-bonds with N‚‚‚H distances of 2.02(8), 2.09(8),
and 2.10(8) Å. In contrast, the singly protonated receptor (5H+)
accepts two H-bonds from the bound H2O and donates one
H-bond to it. The crystal structure (Figure 1) shows that one of
the three hydrogens involved in the H-bonding is located closer
to the pyrazole nitrogen (0.92(3) Å) than to the water oxygen
(1.78(3) Å) while the other two are positioned closer to the water
oxygen (0.88(3), 0.84(3) Å) than to the pyrazole nitrogens
(2.01(3), 2.07(3) Å). The reason the positions of the three
hydrogens are not averaged or disordered appears to be due to
the positioning of the counteranion close to the protonated
pyrazole. Another difference between the bound NH4+ and the
bound H2O is that the former can donate a H-bond to the
counteranion but the latter can only accept a H-bond. Consistent
with this interpretation, the perchlorate counteranion is far
removed from the bound H2O (Figure 1: O‚‚‚O distance of
5.47(1) Å) while the pentafluorophosphate counteranion is
within H-bonding distance of the bound NH4+ (Figure 2: N‚‚‚F
distance of 3.18(1) Å). This difference in the H-bonding property
Figure 3. The ISE responses of (a) 6 and (b) nonctin for NH4+ at various
pH values (7.2, 8.0, and 9.0).
of the receptor and the cationic center is 2.97 and 2.81 Å for
NH4+ and K+, respectively. Four of the nonactin ether oxygens
+
form H-bonds with NH4 while the four carbonyl oxygens
appear to stabilize the cation by dipole interactions. In contrast,
all eight oxygens of nonactin coordinate to K+. The high
sensitivity of nonactin for binding NH4 and K+ is likely due
+
to the multidentate nature of the receptor and the low selectivity
due to the comparable size and charge of the two cations.
It is well-known that K+ prefers to be coordinated by six or
more atoms.14 Thus, lowering the number of coordinating groups
from eight in nonactin to three in 5 has led to a receptor with
+
of the bound NH4 and the bound H2O could perhaps be used
+
to develop receptors that are highly selective for binding NH4
over H2O. Finally, cation-π interaction17 between the benzene
ring of the receptor and the guest molecule may be weaker in
+
the H2O bound complex relative to that in the NH4 bound
greater selectivity for binding NH4 over K+. However, this
+
complex since the bound water oxygen is not positively charged
as in the bound ammonium nitrogen.
has also resulted in significantly lowering the sensitivity of the
receptor for binding NH4+. One reason for the lowered
sensitivity may be due to binding of water to the singly
protonated form of the receptor. Figure 1 shows the crystal
structure of H2O bound to 5H+ (5H+‚H2O, Scheme 1). This
Two simple approaches can be used to lower the binding of
water and thus increase the sensitivity of the receptor. First,
placing electron withdrawing groups on the pyrazole rings of
+
+
structure is analogous to that of NH4 bound to 5 (5‚NH4
,
(15) Iverson, D. J.; Hunter, G.; Blount, J. F.; Damewood, J. R.; Mislow, K. J.
J. Am. Chem. Soc. 1981, 103, 6073-6083.
Scheme 1).8 The three pyrazole rings converge to form three
(16) Bisson, A. P.; Lynch, V. M.; Monahan, M.-C. K.; Anslyn, E. V. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2340-2342.
(14) Wilkins, G.; Gillard, R. D.; McCleverty, J. A. ComprehensiVe Coordination
Chemistry; Pergamon Press: New York, 1987; Vol. 3, p 3.
(17) Dougherty, D. A. Science 1996, 271, 163-168.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5377

A R T I C L E S
Chin et al.
Table 4. Selectivities and Sensitivities of Ionophores for Binding
be less basic than that in 5 since the pKb of 1,3,5-trimethyl-4-
bromopyrazole (12.25) is significantly higher than that of 1,3,5-
trimethylpyrazole (10.26).18 The more basic receptor should
protonate more easily to form the water binding receptor (5H+,
Scheme 1). Since NH4 competes for the water binding site,
the sensitivity of the more basic receptor for binding the cation
NH4 over K+ at pH 8.0
+
+
+
ionophores
selectivity (logKNH
)
detection limit (M)
/K
4
5
6
7
-2.8
-2.3
-1.7
-1.3
1.0 × 10-4
2.5 × 10-5
2.0 × 10-4
2.2 × 10-5
+
nonactin
is expected to be lower (Scheme 1). Indeed, 6 is more sensitive
+
than 5 for binding NH4 (Table 4). While 5 is less sensitive
+
the receptor (6) should reduce its basicity and decrease the
concentration of the singly protonated form of the receptor
relative to the neutral form at a given solution pH (Scheme 1).
than 6, it is somewhat more selective than 6 for binding NH4
over K+ (Table 4). Thus, it appears that the electron withdrawing
group (Br) in 6 lowers the strength of the N-H-N hydrogen
bonds more than the strength of the K-N bonds. In agreement
with this experimental result, ab initio computation shows that
+
Such electronic tuning should favor the binding of NH4 over
H2O. Second, increasing the solution pH should also decrease
the singly protonated form of the receptor and increase the
sensitivity of the receptor for binding NH4+ over H2O. However,
+
5 is more selective than 6 for binding NH4 over K+.19
It is also interesting to compare the sensitivities and selec-
+
the solution pH should be kept below the pKa of NH4 since
tivitites of 5 and 7 for binding NH4+. While 5 is more selective
ISEs are not responsive to NH3.
+
than 7 for binding NH4 over K+, the two receptors have
Figure 3a shows that with increasing solution pH, the
sensitivity of 6 based ISEs approaches that of nonactin based
ISEs (Figure 3b). The sensitivity of nonactin does not change
significantly with changing pH. Receptor 6 is about equally
sensitive as nonactin at pH 9 but the former is about 10 times
more selective than the latter for binding NH4+ over K+ (Table
4). It is interesting that the combination of three H-bonds and
a cation-π interaction is enough for 6 to have comparable
sensitivity as nonactin.
+
comparable sensitivities for binding NH4 (Table 4). As
mentioned above for 6, the three methyl groups on the 3-position
of the pyrazole rings in 5 are expected to increase the selectivity
of the receptor for binding NH4 over K+ by blocking any
+
ligands from coordinating to K+ from the top. The three methyl
groups on the 5-position of the pyrazole rings in 5 are expected
+
to increase the sensitivity of the receptor for binding NH4 by
converging the three imine nitrogens. However, the basicity of
the pyrazoles in 5 is greater than that in 7 and this effect is
expected to make 7 more sensitive (Scheme 1). Thus the
electronic effect of the methyl groups in 5 is to decrease the
sensitivity of the receptor while the steric effect of the methyl
groups on the 5-position of the pyrazoles is to increase the
sensitivity of the receptor and the two effects appear to cancel
each other (Table 4).
To dissect the steric and electronic effects of the methyl and
bromo groups in 6, we compared the selectivity and sensitivity
of 6 to those of 5 and 7 for binding NH4+. The basicity of the
pyrazole groups in 6 is expected to be close to that in 7 since
the pKb of 1-methylpyrazole (11.94) is comparable to that of
1,3,5-trimethyl-4-bromopyrazole (12.25).18 Thus the electronic
effect of the pyrazoles in 7 should be similar to that in 6 and
the steric effects of the methyl groups in 6 can be studied in
isolation. The three methyl groups on the 3-position of the
pyrazole rings in 6 were intended to increase the selectivity of
the receptor for binding NH4+ over K+ by blocking any ligands
from coordinating to K+ from the top. The crystal structure
In summary, an artificial receptor (6) based ISE is as sensitive
(at pH 9) and 10 times more selective than nonactin based ISEs
for sensing NH4+ over K+. Steric and electronic effects, as well
as H-bonding and cation-π interactions, were taken into
consideration in designing the receptor. The three pyrazole
+
+
(Figure 2) of NH4 bound to 6 reveals that the methyl groups
groups in 6 were designed to accept three H-bonds from NH4
.
on the 3-position of the pyrazoles shield the cation such that it
would be difficult to form a 2:1 complex between the receptor
and K+. This strategy appears to have worked since the
selectivity of 6 is significanctly greater than that of 7 for binding
The three ethyl groups in 6 were designed to direct the three
pyrazole rings to the same side of the benzene ring by steric
interactions.15 The three methyl groups on the 5-position of the
pyrazole rings were designed to converge the imine nitrogen
+
NH4 over K+ (Table 5). The three methyl groups on the
+
atoms for binding NH4 by steric interactions and the three
5-position of the pyrazole rings in 6 were placed to direct the
three imine nitrogens inward for binding NH4+. This preor-
ganization effect is expected to increase the sensitivity of the
receptor by increasing the equilibrium constant for binding
NH4+. This strategy also appears to have worked since the
methyl groups on the 3-position of the pyrazole rings were
designed to block ligands from coordinating to the top. The
three electron withdrawing groups (Br) on the receptor were
placed to increase the sensitivity of the receptor for binding
+
NH4 by decreasing the water binding form of the receptor
+
sensitivity of 6 for binding NH4 is significantly greater than
that of 7 (Table 4).
(6H+). Finally, the benzene ring of the receptor is expected to
+
stabilize NH4 by cation-π interactions. The crystal data and
Just as we have been able to isolate the steric effects of the
methyl groups in 6 it is possible to isolate the electronic effect
of the bromo groups in the receptor. The steric effects of the
methyl groups in receptors 6 and 5 should be similar but the
electronic effects of the pyrazole groups in the two receptors
are significantly different. The pyrazoles in 6 are expected to
(19) Ab initio calculations were performed using the GAMESS program.
Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M.
S.; Jensen, J. H.; Koseki, J. H.; Matsunaga, N.; Nguyen, K. A.; Su, S. J.;
Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993,
14, 1347-1363. The geometries were fully optimized under the C3
symmetry condition at the RHF/3-21G* level. The binding energies of the
complexes were corrected for basis set superposition error (BSSE). BSSE
is calculated by using the counterpoise method with a modification that
takes into account the change in geometries upon complexation. Boys, S.
F.; Bernaradi, F. Mol. Phys. 1970, 19, 553-566. Nagy, P. I.; Smith, D.
A.; Alagona, G.; Ghio, C. J. Phys. Chem. 1994, 98, 486-493.
(18) Catalan, J.; Abboud, L. M.; Elguero, J. AdV. Heterocycl. Chem. 1987, 41,
187-274.
9
5378 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

Electronic and Steric Effects in Ammonium Receptors
A R T I C L E S
ISE data support our design rational and can be used to
quantitatively dissect the electronic and steric effects of the
bromo and methyl substitutents on the selectivity and sensitivity
of 6 for binding NH4+. The crystallographic investigation
provides valuable insight into the origin of selective binding of
the receptor to NH4+ over H3O+. This receptor with its simple
Acknowledgment. We gratefully acknowledge the Creative
Research Initiative Program of the Korean Ministry of Science
and Technology and the Natural Sciences and Engineering
Research Council of Canada for support of this work and the
BK21 Program of the Korean Ministry of Education for a
graduate fellowship to J.Oh. We also thank J. Heo for the crystal
structures shown in Figures 1 and 2.
+
platform allows detailed understanding of NH4 recognition,
and it is ideally set up for molecular mechanics and ab initio
computations.15,19 Fundamental information gained from the
studies of NH4+ recognition may be helpful not only for sensing
this cation but also for sensing other structurally related and
biologically interesting compounds such as amino acids and the
dopamine class of neurotransmitters.
Supporting Information Available: Crystallographic infor-
-
-
mation for [5H+‚H2O] ClO4 and [6NH4+] PF6 (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA0174175
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5379