Design and synthesis of a more highly selective ammonium ionophore than nonactin and its application as an ion-sensing component for an ion-selective electrode

Article abstract of DOI:10.1021/ac9911241

A novel ammonium ionophore, which exhibits superior NH₄⁺ selectivity compared with that of the natural antibiotic nonactin, was successfully designed and synthesized based on a 19-membered crown compound (TD19C6) having three decalin

Full text of DOI:10.1021/ac9911241

Anal. Chem. 2000, 72, 2200-2205
Design and Synthesis of a More Highly Selective
Ammonium Ionophore Than Nonactin and Its
Application as an Ion-Sensing Component for an
Ion-Selective Electrode
Koji Suzuki,*,†,‡ Dwi Siswanta,† Takeshi Otsuka,† Tsuyoshi Amano,† Takafumi Ikeda,†
Hideaki Hisamoto,†,| Ryoko Yoshihara,§ and Shigeru Ohba§
Department of Applied Chemistry and Department of Chemistry, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama
223-8522, Japan, and Kanagawa Academy of Science and Technology (KAST), KSP West-614, 3-2-1 Sakato,
Kawasaki 213-0012, Japan
ammonia.18,19 Until recently, the only available practically used
ammonium ionophore was nonactin, a natural product, which was
first employed as the ion-sensing component for an ammonium
ion-selective electrode in 1970.20 For many other ions, following
the use of natural products as ionophores, many researchers
designed and synthesized highly selective synthetic ionophores
for an objective ion, but there are no practically successful
ionophores for NH4+ except nonactin.21,22 Previously, we reported
the synthesis of novel ammonium ionophores based on glycol
benzyl ether derivatives,9,23 which showed a relatively high
ammonium ion selectivity toward K+, but their NH4+-to-Na+ ratios
were not sufficient for practical use when utilized as the ion-
sensing component for an ammonium ion-selective electrode. A
similar result was obtained with a pyrazol-containing crown ether
derivative by Moriuchi-Kawakami et al.10 Here we report the first
successfully designed and synthesized ammonium ionophore
more highly selective than nonactin.
A novel ammonium ionophore, which exhibits superior
NH4 selectivity compared with that of the natural anti-
biotic nonactin, was successfully designed and synthe-
+
sized based on
a 1 9 -membered crown compound
(TD1 9 C6 ) having three decalino subunits in the macro-
cyclic system. This bulky decalino subunit is effective for
(1 ) increasing the structural rigidity of the cyclic com-
pound, (2 ) introducing the “block-wall effect”, which
prevents forming a complex with a large ion, and (3 )
increasing the lipophilicity of the ionophore molecule. In
the ammonium ionophore design, the first factor contrib-
utes to increasing the NH4 + selectivity relative to smaller
ions such as Li+, Na+, or even the closest size, K+, and
+
the second factor increases the NH4 selectivity over
larger ions such as Rb+ and Cs+. The X-ray structural
analysis proved that TD19C6 forms a size-fit complex with
+
NH4 in its crown ring cavity. As an application of this
Most of the useful ionophores for alkali and alkaline earth
metal ions are based on a crown ether cyclic structure; therefore,
ionophore, an ion sensor (ion-selective electrode) was
prepared, which exhibited NH4 + to K+ and Na+ selectivity
of 10 and 3000 times, respectively. This electrode showed
a better performance compared to the electrode based on
nonactin, which is the only ammonium ionophore pres-
ently used in practical applications.
(8) Knoll, M.; Cammann, K.; Dumschat, C.; Sunder-meier, C.; Eschold, J. Sens.
Actuators 1 9 9 4 , B18-19, 51.
(9) Siswanta, D.; Hisamoto, H.; Tohma, H.; Yamamoto, N.; Suzuki,K. Chem.
Lett. 1 9 9 4 , 945.
(10) Moriuchi-Kawakami, T.; Nakazawa, S.; Ota, M.; Nishihara, M.; Hayashi, H.;
Shibuyani, Y.; Shono, T. Anal. Sci. 1 9 8 8 , 14, 1065.
(11) Schindler, J. G. Eur. J. Clin. Chem. Clin. Biochem. 1 9 9 4 , 32, 145.
(12) Alegret, S.; Bartoli, J.; Jimenez, C.; Martinez-Fabregas, E.; Martorell, D.;
Valdes-Perezgasga, F. Sens. Actuators 1 9 9 3 , B15-16, 453.
(13) Vaillo, E.; Walder, P.; Spichiger, U. E. Anal. Methods Instrum. 1 9 9 5 , 2,
145.
The availability of a highly selective ammonium ionophore is
crucial for the preparation of ammonium ion chemical sensors
and biosensors used for the direct measurement of the ammonium
ion1-10 as well as ammonium-derived urea,11-15 amines,16,17 or
(14) Borchardt, M.; Dumschat, C.; Cammann, K. Sens. Actuators 1 9 9 5 , B24-
25, 721.
(15) Stamm, C.; Seiler, K.; W. Simon Anal. Chim. Acta 1 9 9 3 , 282, 239.
(16) Odashima, K.; Yagi, K.; Tohda, K.; Umezawa, Y. Anal. Chem. 1 9 9 3 , 65,
1074.
(17) Chan, W. H.; Lee, A. W. M.; Wang, K. Analyst 1 9 9 4 , 119, 2809.
(18) Ozawa, S.; Hauser, P. C.; Seiler, K.; Tan, S. S. S.; Morf, W. E.; Simon, W.
Anal. Chem. 1 9 9 1 , 53, 640.
(19) West, S. J.; Ozawa, S.; Seiler, K.; Tan, S. S. S.; Simon, W. Anal. Chem. 1 9 9 2 ,
54, 633.
(20) Scholer, R. P.; Simon, W. Chimia 1 9 7 0 , 24, 372-374.
(21) Umezawa, Y. Handbook of Ion-Selective Electrodes: Selectivity Coefficients;
CRC Press: Boca Raton, FL, 1990.
(22) Buhlmann, P.; Bakker, E.; Pretch, E. Chem. Rev. 1 9 9 8 , 98, 1593.
(23) Suzuki, K. J. Synth. Org. Chem. Jpn. 1 9 9 8 , 56, 291.
† Department of Applied Chemistry, Keio University.
‡ Kanagawa Academy of Science and Technology (KAST).
§ Department of Chemistry, Keio University.
| Present address: Department of Applied Chemistry, Graduate School of
Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8566.
(1) Ghauri, M. S.; Thomas, J. D. R. Analyst 1 9 9 4 , 119, 2323.
(2) Thanei-Wyss, U.; Morf, W. E.; Lienemann, P.; Stefanac, Z.; Mostert, I.; Dorig,
R.; Dohner, R. E.; Simon, W. Mikrochim. Acta 1 9 8 3 , Ill, 135.
(3) Ma, S. C.; Chaniotakis, N. A.; Meyerhoff, M. E. Anal. Chem.1 9 8 8 , 60, 2293.
(4) Beer, D. De.; van den Heuvel, J. C. Talanta 1 9 8 8 , 35, 728.
(5) Fresser, F.; Moser, H.; Nair, M. J. Exp. Biol. 1 9 9 1 , 147, 227.
(6) Henriksen, G. H.; Bloom, A. J.; Spanswick, R. M. Plant. Physiol. 1 9 9 0 93,
271.
(7) Seiler, K.; Morf, W. E.; Rusterholz, B.; Simon, W. Anal. Sci. 1 9 8 9 . 5, 557.
2200 Analytical Chemistry, Vol. 72, No. 10, May 15, 2000
10.1021/ac9911241 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/06/2000

Figure 1. Effective molecular structure design for ammonium ionophore (target ion, NH4+).
Potassium ion and ammonium ion have a similar ionic size,
but considering the hydrogen-bonding distance of K+‚‚‚O and
+
N-H‚‚‚O, the latter bond is slightly longer. Consequently, NH4
has a slightly larger ionic size than that of K+. Because 18-crown-6
is the best fit for K+ in its cavity, we then designed a larger size
of crown ether using 19-crown-6 as the backbone. Based on the
molecular model shown in Figure 1, three ionophores using a
19-crown-6 structure shown in Figure 2 were synthesized. As a
result, TD19C6, which possessed three decalino subunits in the
19-membered crown ring, exhibited a high ammonium ion
selectivity toward other alkali metal and alkaline earth metal ions.
This ammonium ion selectivity is better than that of nonactin.
EXPERIMENTAL SECTION
Reagents. The highest grade commercially available reagents
were used for the syntheses of the new compounds and the
preparation of the aqueous test electrolytes. The distilled and
deionized water used had a resistivity of greater than 1.5 × 107 Ω
cm at 25 °C. The electrode membrane solvent bis(1-butylpentyl)
adipate (BBPA) was purchased from Fluka AG, Buchs, Switzer-
land. Poly(vinyl chloride) (PVC, high molecular weight type) used
as the electrode membrane material was obtained from Sigma,
Chemical Co. (St. Louis, MO).
Synthesis of 1 9 -Crown-6 Derivatives. 2 ,5 ,1 2 ,1 5 ,2 2 ,2 6 -
Hexaoxaheptacyclo[2 5 .4 .4 .4 6 .1 1 4 1 6 .2 1 .0 1 .2 7 .0 6 .1 1 .0 1 6 .2 1 ]-
tritetracontane (TD19C6). A solution of cis-1,6-dihydroxybicyclo-
[4.4.0]decane (10 g, 58.7 mmol), malonaldehyde bis(dimethylacetal)
(4.778 g, 29.1 mmol), and p-toluenesulfonic acid (0.477 g) in
benzene (50 mL) was refluxed for 2 h. The reaction mixture was
then evaporated and partitioned three times with chloroform. The
chloroform was then evaporated. The residue was dissolved in a
small amount of chloroform (20 mL), methanol (20 mL) was then
added, and the solution cooled at 0 °C for 2 h. The resulting white
crystals were filtrated, separated, and identified as 12,12′-methyl-
enedi-11,13-dioxatricyclo[4.4.3.01.6]tridecane (product 1 , 8.74 g,
79.8%).
Figure 2. Chemical structures of the newly synthesized ammonium
ionophores (TD19C6, DD19C6, TTM19C6) which have a 19-
membered-crown ring with three or two bulky block subunits and
nonactin.
a major obstacle exists in employing a normal crown ether for
designing an ammonium ionophore because the ammonium ion
has an ionic size similar to that of K+.23 We now introduce two
major and unique factors to be considered when designing an
ammonium ionophore based on a crown ether basic structure,
i.e., introducing a rigid frame and “block-walls” in the macrocyclic
structure, preventing the formation of both a wrapping complex
of the crown ether with a smaller ion and a sandwich complex
with a larger ion, as demonstrated in Figure 1. We selected a
decalino subunit (derived from decalin diol 1 ) to satisfy both the
requirements of rigidity and the “block-wall effect” and incorporate
it into the macrocyclic system. For a comparative study, a
tetramethylethano-subunit (derived from pinacol) was also used
as the “block” subunit, and highly lithium and sodium ion-selective
ionophores were successfully obtained.24,25
A solution of lithium aluminum hydride (1.586 g, 41.8 mmol)
in ether was added to the solution of aluminum chloride (22.29
g, 167.2 mmol) in ether, and the mixture stirred in a water bath
for 30 min. Product 1 was then added and stirred for 3 h. After
(24) Suzuki, K.; Yamada, H.; Sato, K.; Watanabe, K.; Hisamoto, H.; Tobe, Y.;
Kobiro, K. Anal. Chem. 1 9 9 3 , 65, 3404.
(25) Suzuki, K.; Sato, K.; Hisamoto, H.; Siswanta, D.; Hayashi, K.; Kasahara, N.;
Watanabe, K.; Yamamoto, N.; Sasakura, H. Anal. Chem. 1 9 9 6 , 68, 208.
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000 2201

the addition of water followed by the addition of a 10% H2SO4
solution, the solvent was decanted and the water phase was
extracted three times with ethyl acetate. The organic solvent was
evaporated, and the obtained residue was purified by silica gel
column chromatography with hexane-ethyl acetate (4:1) to yield
6,6′-(propylenedioxy)di-cis-1-hydroxydecane (product 2 , 7.13 g,
85.2%) as a white solid.
organic phase, the residue was dissolved in chloroform and
washed with HCl solution (pH 1) and brine. The chloroform was
evaporated, and the residue was purified by silica gel chroma-
tography (hexanes-ethyl acetate 4:1) to give 7 (2.03 g, 19.9%).
NaH (200 mg, 5 mmol) was added to the solution of product
2 (267.5 mg, 0.87 mmol) in THF (20 mL) and stirred for 1 h at 40
°C. A solution of 7 (267.5 mg, 0.87 mmol) in 30 mL of THF was
then added to this mixture. The mixture was stirred at 70 °C for
20 h. A small amount of methanol (10 mL) was added to quench
the excess NaH. After evaporation of the THF, the residue was
extracted twice with ethyl acetate. The organic phase was then
evaporated, and the obtained residue was purified by silica gel
column chromatography (hexane-ethyl acetate 3:1) and further
purification using HPLC with acetone as the eluent to yield
DD19C6 (12 mg, 2.8%) as a colorless oil:1H NMR (270 MHz,
CDCl3) δ 1.18-2.18 (m, 34H, CH2), 3.20-4.05 (m, 16 H, OCH2).
Anal. Calcd for C29H50O6(494.36): C, 70.41; H, 10.19. Found: C,
70.02; H, 10.29.
Chloroacetic acid (5.76 g, 60.93 mmol) was added to a mixture
of cis-1,6-dihydroxybicyclo[4.4.0]decane (2.0 g, 25.39 mmol) and
NaH (4.06 g, 101.5 mmol) in 100 mL of THF and the resultant
mixture refluxed for 16 h. After the addition of methanol (10 mL),
the solvent was evaporated and 50 mL 1 M HCl was added to the
residue and partitioned three times with chloroform. The organic
layer was dried over Na2SO4 and the solvent evaporated to obtain
cis-1,6-di(3′-oxapropionic acid)bicyclo[4.4.0]decane (3 ). Product
3 was then dissolved in 100 mL of ethanol-benzene (1:1), and
3A molecular sieves (50 mg) and p-toluensulfonic acid (0.13 g)
were added to the solution. The reaction mixture was refluxed at
90 °C for 17 h, and the organic solvent was then evaporated. The
residue was dissolved in 100 mL of ethyl acetate and washed with
aqueous NaHCO3 and brine. The organic phase was evaporated,
and the residue was purified by silica gel column chromatography
(hexanes-ethyl acetate 2:1) to obtaincis-1,6-di(3′-oxaethoxypropan-
oyl)bicyclo[4.4.0]decane (4 , 2.8 g, 69.6%).
2 ,2 ,3 ,3 ,8 ,8 ,9 ,9 ,1 4 ,1 4 ,1 5 ,1 5 -Dodecamethyl-1 ,4 ,7 ,1 0 ,-
1 3 ,1 6 -hexaoxacyclononadecane (TTM1 9 C6 ). To a solution of
pinacol (7.27 g, 61.5 mmol) and malonaldehyde bis(dimethylac-
etal) (5.0 g, 30.45 mmol) in benzene (150 mL) was added a
catalytic amount of p-toluenesulfonic acid monohydrate (100 mg),
and the resultant mixture was stirred at 90 °C for 2 h. The reaction
mixture was then evaporated and partitioned three times with
chloroform. The organic phase was evaporated, and the residue
was purified by silica gel column chromatography with hexane-
ethyl acetate (4:1) to yield 2,2-methylenedi-4,4,5,5-tetramethyl-1,3-
dioxolane (product 8 , 7.87 g, 95.0%) as white crystals.
Product 4 (3.2 g, 9.34 mmol) was added to a solution of LiAlH4
(1.03 g, 27.79 mmol) in 30 mL of THF and refluxed for 25 h. After
the addition of methanol (20 mL), the organic solvent was
evaporated and the residue was partitioned three times with ethyl
acetate. After evaporation of ethyl acetate, the resulting residue
was purified by silica gel column chromatography to give cis-1,6-
di-(2′-hidroxyethoxy)bicyclo[4.4.0]decane (5 , 0.7 g, 28.99%).
Product 5 (0.7 g, 2.71 mmol) was dissolved in 10 mL THF,
pyridine (20 mL) was then added, and the mixture was stirred on
an ice bath for 45 min. A solution of tosyl chloride (1.29 g, 6.77
mmol) in 30 mL THF was then slowly added, and the reaction
mixture was stirred for 5 h. After evaporation of the organic phase,
the residue was dissolved in chloroform and washed with HCl
solution (pH 1) and brine. The chloroform was evaporated and
the residue was purified by silica gel chromatography (hexanes-
ethyl acetate 4:1) to give cis-1,6-di-(2′-hidroxyethoxytosylate)-
bicyclo[4.4.0]decane (6 , 0.85 g, 55%).
A solution of lithium aluminum hydride (752 mg, 19.8 mmol)
in 10 mL of ether was added to the solution of aluminum chloride
(10.57 g, 79.3 mmol) in 130 mL of ether, and the resultant mixture
stirred in a water bath for 30 min. The product 8 (3.0 g, 11 mmol)
was then added and stirred for 3 h. After the addition of water
followed by addition of 10% H2SO4 solution (25 mL), the organic
phase was decanted and the water phase was extracted three times
with ethyl acetate. The organic phase was collected and evapo-
rated. The obtained residue was purified by silica gel column
chromatography with hexane-ethyl acetate (1:1) as the eluent
to yield 2,3,3,9,9,10-hexamethyl-4,8-dioxa-2,10-undecanediol (prod-
uct 9 , 2.46 g, 80.9%) as a white solid.
NaH (141 mg, 11 mmol) was added to the solution of 2 (335.8
mg, 0.88 mmol) in THF (20 mL) and the resultant mixture stirred
for 1 h at 40 °C. A solution of 6 (500 mg, 0.88 mmol) in 30 mL of
THF was then added to this mixture. The mixture was then stirred
at 70 °C for 6 days. A small amount of methanol (10 mL) was
added to quench the excess NaH. After evaporation of the THF,
the residue was extracted twice with ethyl acetate. The organic
phase was evaporated, and the obtained residue was purified by
silica gel column chromatography (hexane-ethyl yield TD19C6
(9 mg, 1.7%) as white crystals. 1H NMR (270 MHz, CDCl3) δ 1.25-
2.35 (m, 50H, CH2), 3.3-4.1 (m, 12 H, OCH2). Anal. Calcd for
C37H62O6(602.89): C, 73.71; H, 10.36. Found: C, 73.32; H, 10.56.
Chloroacetic acid (5.76 g, 60.93 mmol) was added to a mixture
of pinacol (3.0 g, 25.39 mmol) and NaH (4.06 g, 101.5 mmol) in
100 mL of THF and refluxed for 16 h. After the addition of
methanol (10 mL), the solvent was evaporated and then 50 mL of
1 M HCl was added to the residue and partitioned three times
with chloroform. The organic phase was dried over Na2SO4 and
evaporated to obtain 3,3,4,4-tetramethyl-2,5-dioxa-1,6-hexanedioic
acid (1 0 ). The product 1 0 was then dissolved in 100 mL of
ethanol-benzene (1:1), and molecular sieves 3A (50 mg) and
p-toluenesulfonic acid (0.13 g) were added to the solution. The
reaction mixture was refluxed at 90 °C for 17 h, and then the
organic solvent was evaporated. The residue was dissolved in 100
mL of ethyl acetate and washed with aqueous NaHCO3 and brine.
The solvent was evaporated, and the residue was purified by silica
gel column chromatography (hexanes-ethyl acetate 2:1) to obtain
3,3,4,4-tetramethyl-2,5-dioxa-1,6-hexanedioic acid diethyl ester (11,
2.33 g, 31.6%).
(2 ,6 ,1 3 ,1 6 ,1 9 ,2 2 -Hexaoxapentacyclo[2 1 .4 .4 .1 .2 3 4 7 .1 2
-
0 1 .2 3 .0 7 .1 2 ]pentatriacontane (DD1 9 C6 ). Mesyl chloride (8.01
g, 69.9 mmol) was added to a solution of triethylene glycol (5.0
g, 33.3 mmol) in 50 mL of stirred pyridine in an ice bath. This
reaction mixture was stirred for 5 h. After evaporation of the
2202 Analytical Chemistry, Vol. 72, No. 10, May 15, 2000

Product 1 1 (3.2 g, 9.34 mmol) was added to a solution of
LiAlH4 (1.03 g, 27.79 mmol) in 30 mL of THF and refluxed for 25
h. After the addition of methanol (20 mL), the organic solvent
was evaporated and the residue was partitioned three times with
ethyl acetate. After evaporation of the ethyl acetate, the resulting
residue was purified by silica gel column chromatography to give
4,4,5,5-tetramethyl-3,6-dioxa-1,8-octanediol (1 2 , 0.82 g, 49.25%).
Product 1 2 (0.82 g, 1.5 mmol) was dissolved in 10 mL of THF,
pyridine (20 mL) was then added, and the mixture was stirred in
an ice bath for 45 min. A solution of tosyl chloride (0.71 g, 3.74
mmol) in 30 mL of THF was then added slowly, and the reaction
mixture was stirred for 5 h. After evaporation of the organic phase,
the residue was dissolved in chloroform and washed with HCl
solution (pH 1) and brine. The chloroform was evaporated, and
the residue was purified by silica gel chromatography (hexanes-
ethyl acetate 4:1) to give 1 3 (1.19 g, 58.57%).
NaH (135 mg, 10.6 mmol) was added to the solution of product
9 (238 mg, 0.86 mmol) in THF (20 mL), and the resultant mixture
stirred for 1 h at 40 °C. A solution of 1 3 (450 mg, 0.88 mmol) in
30 mL of THF was then added. The mixture was then stirred at
70 °C for 3 days. A small amount of methanol (10 mL) was added
to quench the excess NaH. After evaporation of the THF, the
residue was extracted twice with ethyl acetate. The organic phase
was then evaporated, and the obtained residue was purified by
silica gel column chromatography (hexane-ethyl acetate 4:1) and
further purification using HPLC with acetone as the eluent to yield
TTM19C6 (18.2 mg, 4.7%) as a colorless oil: 1H NMR (270 MHz,
CDCl3) δ 1.14-1.15 (d, J ) 4.34 Hz, 36H, CH3), 1.63-1.69 (t, 2 H,
CH2), 3.37-3.57 (m, 12H, OCH2). Anal. Calcd for C25H50O6(446.67):
C, 67.23; H, 11.28. Found: C, 67.02; H, 11.49.
The lipophilicity (log po/ w, the partition coefficient of oil and
water) of the newly synthesized ionophores was measured
according to a previously reported procedure.26
X-ray Crystal Structure Analysis. The X-ray crystallographic
analysis for TD19C6 was performed using an AFC-7R analyzer
(Rigaku-Denki Co., Ltd., Tokyo, Japan). The single-crystal sample
was obtained by recrystallization of the TD19C6-NH4+SCN-
complex with acetone. The detailed crystallographic data for
TD19C6 are described in the Supporting Information.
Electrode Membrane P reparation and EMF Measure-
ments. For the potentiometric studies, the newly synthesized
ionophores (TD19C6, DD19C6, TTM19C6) were then incorpo-
rated in a polymeric membrane electrode using the KTCPB
previously reported procedures. The membrane composition was
3 wt % ionophore, 10 mol % (relative to the ionophore) anionic
additive (tetrakis(p-chlorophenyl) borate potassium salt, KTCPB),
67 wt % membrane solvent, BBPA), and 30 wt % PVC (high
molecular weight). The membrane thickness was ∼100 µm. A
6-mm-diameter circle was cut from a prepared membrane and
placed on the tip of the ion-selective electrode body assembly.
The prepared electrodes were immersed in 0.1 M NH4Cl solution
for over 24 h for preconditioning before use. The electrode
response potential (emf) measurements were performed according
to the reported procedure at 25 ( 0.5 °C using the electrochemical
cell system.27
Ag|AgCl|3 M KCl||test solution|membrane|
0.1 M NH4Cl|AgCl|Ag
All test solutions were made from the chloride salt without any
pot
pH-adjusting buffer reagent. The selectivity coefficients, Kij
,
where i is the primary ion (NH4+) and j is the interfering ion,
were calculated from the response potentials in an alkali metal or
alkaline earth metal chloride solution using the separate solution
method (SSM, i ) j ) 0.1 M) according to the recommendations
of the IUPAC and JIS.28,29
RESULTS AND DISCUSSION
As described in the introduction, the aim of the present
investigation was to obtain a higher NH4+-selective ionophore
compared to that of nonactin that was designed on the basis of a
molecular model involving a bulky block-wall and rigid subunits,
+
as shown in Figure 1. The structures of our synthesized NH4
ionophore (Figure 2) are TD19C6, DD19C6, and TTM19C6
containing two or three decalino or tetramethylethano subunits
as the block-wall subunits in their basic skeleton of the 19
-membered crown 6.
Regarding the synthesis of these ionophores, the overall
synthetic yields of these 19-membered cyclic compounds were
low because the reaction with the highly hindered tertiary alcohols
were utilized in their synthesis. However, implementing efficient
procedures for the synthesis of important decalino units of 6,6′-
(propylenedioxy)di-cis-1-hydroxydecane (2 ) by Sachleben et al.30
and product 5 by Kobiro et al.,31 then the only low-yield reaction
(∼10% yield) was the final coupling using product 2 and the
ditosylate of product 5 . The synthesis work involving the tetram-
ethyl subunit for the preparation of TTM19C6 gave a higher
overall yield, but in our experience, the product, which has a
higher final coupling yield, showed a lower block effectivity and
structural rigidity compared to the decaline diol unit. Furthermore,
the tetramethyl subunit exhibited a lower contribution to the
lipophilicity of the ionophore, which is an important parameter
for application of the ionophore as a sensory molecule in an ion-
selective electrode membrane. The typical lipophilicity parameter,
log po/ w, of the ionophores is 13.5 ( 0.7 for TD19C6 and 12.2 (
0.6 for DD19C6, which possess three and two decalino subunits,
respectively, while the log po/ w of TTM19C6, which has three
tetramethyl subunits, is 7.9 ( 0.6 and that for nonactin is 5.8 (
0.4. Consequently, the decalino subunits incorporated in the 19-
crown-6 backbone structure (TD19C6) finally have the three roles
of rigidity, “block-wall effect”, and added lipophilicity to the
ionophore molecule.
+
The selectivity coefficients of the NH4 electrodes based on
TD19C6, DD19C6, TTM19C6, and nonactin are shown in Figure
3. The electrode based on TD19C6 exhibited high selectivity
relative to alkali metal and alkaline earth metal ions, especially to
K+ and Na+, which are the major sources of interferences in the
sensing of NH4+. The ionic selectivity pattern was NH4+ > K+
>
(28) IUPAC Recommendation for Nomenclature of Ion-Selective Electrodes. Pure
Appl. Chem. 1 9 7 6 , 48, 29.
(29) JIS K-0122, Japanese Standards Association, Tokyo, 1997.
(30) Sachleben, R. A.; Davis, M. C.; Bruce, J. J.; Ripple, E. S.; Driver, J. L.; Moyer,
B. A. Tetrahedron Lett. 1 9 9 3 , 34, 5373.
(26) Dinten, O.; Spichger, U. E.; Chaniotokis, N.; Gehrig, P.; Rusterholz, B.; Morf,
W. E.; Simon, W. Anal. Chem. 1 9 9 1 , 63, 596.
(27) Suzuki, K.; Tohda, K.; Aruga, H.; Matsuzoe, M.; Inoue, H.; Shirai, T. Anal.
Chem. 1 9 8 8 , 60, 1714.
(31) Kobiro, K.; Ohnishi, K.; Kakiuchi, K.; Tobe, Y.; Odaira, Y. Bull. Chem. Soc.
Jpn. 1 9 8 5 , 58, 1333.
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000 2203

Figure 3. Selectivity coefficients of the ammonium ion-selective electrodes based on TD19C6, DD19C6, TTM19C6, and nonactin. The membrane
compositions for the three ionophores based on 19C6 derivatives were 3 wt % ionophore, 10 mol % (of the ionophore content) KTCPB, 67 wt
% BBPA, and ∼30 wt % PVC. The ion-selectivity factors of the nonactin-based electrode were obtained from ref 9.
Rb+ > Cs+ > Na+ > Mg2+ > Ca2+. When a normal 19-crown-6
without block subunits is used, it would be hard to obtain a high
ability to differentiate NH4+ from K+ and Rb+, because the normal
crown compound is flexible enough to accommodate ions with
similar sizes. A very attractive fact for TD16C5 is that Na+, which
is smaller than NH4+, is strongly rejected. In addition, Cs+, which
is larger compared to NH4+, was also strongly discriminated.
Some typical electrode membrane solvents such as BBPA,
2-nitrophenol octyl ether, (NPOE), and tris(2-ethylhexyl phos-
phate) (TEHP) were used and examined for ion selectivity with
+
the electrode based on TD19C6. The best results in the NH4
selectivity against Na+ and K+ were obtained in the case where
BBPA, which has an low dielectric constant (ꢀ ) 4; where ꢀ is
dielectric constant), was used as the membrane solvent. For the
+
Figure 4. Steric chemical structures of TD19C6-NH4 complex
determined by X-ray analysis.
anionic additive (TCPB anion) composition, the best results for
+
NH4 selectivity were obtained where the TCPB used was 10-
We have successfully designed and synthesized highly selec-
tive Li+ and Na+ ionophores by introducing one and two decalino
subunits in a 14-crown-4 and 16-crown-5, respectively.24,25 In the
present investigation, introducing three decalino subunits in a 19-
crown-6 resulted in a significant breakthrough in designing a
highly selective ammonium ionophore, because K+ and NH4+ are
strikingly difficult to discriminate. In our previous study, a
relatively rigid and blocked podand-type glycol benzyl ether
exhibited a high ammonium-to-K+ selectivity, but it failed to
improve the rejection of Na+.18 The present work has overcome
this problem.
40mol % against the ionophore content in the electrode membrane.
The electrode based on DD19C6, which possesses two decalino
subunits, showed NH4+/ K+ and NH4+/ Na+ selectivities of -0.24
and -2.23, respectively. These selectivity values indicate that
TD19C6 has a higher ring rigidity and block-wall effect than those
with DD19C6. With only two blocking subunits, the triethylene
glycol subunit is an open and flexible segment which reduces the
+
+
NH4 selectivity of DD19C6. The drastic lowering of the NH4
-
to-Na+ selectivity rather than the decreasing of the Cs+ selectivity
should indicate that the decreasing of the structural rigidity of
DD19C6 was more dominant than the decreasing of the block-
wall effect for larger size ions.
The ammonium complex structure determined by three-
dimensional X-ray analysis showed that only three of four
hydrogen atoms of the ammonium ion coordinate with six oxygen
atoms in the crown as shown in Figure 4, thus leaving one
hydrogen atom free in the upward direction relative to the crown
ring plane, which is normally coordinated to a counteranion. The
determined steric structure of TD19C6 proved that the ionophore
The electrode based on TTM19C6, which has three tetram-
ethylethano subunits, exhibited NH4+/ K+ and NH4+/ Na+ selectivi-
ties of -0.66 and -2.82, respectively. Overall, regarding the ionic
selectivity of the electrode based on TD19C6, the tetramethyl-
ethano subunit is inferior in terms of increasing the rigidity of
the crown structure and also in terms of the effectivity in
producing the block-wall effect compared to the decalino subunit.
+
molecule forms a size-fit complex with NH4 in its crown ring
2204 Analytical Chemistry, Vol. 72, No. 10, May 15, 2000

cavity which was a satisfactory result that we expected in the
design stage.
the present report. This molecular design concept opens the way
to obtaining highly analyte-selective ionophores and practically
useful ion sensors.
The NH4+/ K+ and NH4+/ Na+ selectivities of the electrode
based on TD19C6 were -1.00 and -3.52, respectively. Although
this electrode exhibited a similar NH4+/ K+ selectivity compared
with the selectivity of the electrode based on nonactin, it showed
a very drastic improvement in ammonium selectivity over Na+
(∼10 times). The electrode based on TD19C6 showed a response
ACKNOWLEDGMENT
Partial support of this investigation by the Kanagawa Academy
of Science and Technology and the Ministry of Science and
Technology is acknowledged. This study was also partially
supported by a Grant-in-Aid for Scientific Research from the
Ministry of Education, Science, Sports and Culture.
+
to ammonium ions in the activity range of 5 × 10-6-10-1 M NH4
in an almost Nernstian response (the average slope obtained with
five electodes using TD19C6 was 58.1 mV/ activity decade). The
electrode response slope and selectivity were very similar to the
values during the first day of use.
SUPPORTING INFORMATION AVAILABLE
The X-ray crystallographic analysis data for TD19C6. This
material is available free of charge via the Internet at http:/ /
pubs.acs.org.
In conclusion, this is the first case of obtaining a highly
ammonium ion-selective ionophore, TD19C6, which exhibits
+
superior NH4 selectivity compared to that of nonactin. Thus,
TD19C6 is the best ammonium ionophore developed to date. The
designing of an ionophore based on a unique concept that involves
reducing the coordinating ability with interfering ions other than
the objective ion by using “block” subunits was demonstrated in
Received for review September 29, 1999. Accepted
February 8, 2000.
AC9911241
Analytical Chemistry, Vol. 72, No. 10, May 15, 2000 2205